← Lucian Cosinschi

A1 — Foundations, 1903–1980

How pilots were trained before automation forced the question, and how the intellectual machinery of aviation human factors was built

Research stream A1. Evidence memo. Verification tags: [VERIFIED] = source read directly or via authoritative full text; [SECONDARY] = held only through a citing source; [UNVERIFIED] = could not confirm.

0. Summary of the argument

Between 1903 and 1980 aviation trained pilots on an implicit theory that the pilot was a skilled individual — a person with the right aptitude, whose hands and judgement could be measured, selected for, and drilled. Every institution built in that period followed from the theory: selection batteries, elimination (“washout”) rates, the check ride, the type rating, the proficiency check. The theory worked well enough that by 1970 the residual accident causes were no longer things individuals did badly. They were things crews failed to do together, and things designers had made easy to do wrongly.

The correction did not come from training research. It came from three other places: (a) a design critique launched in 1943–1951 that reclassified “pilot error” as designer error; (b) an accident record between 1972 and 1978 in which technically excellent crews destroyed serviceable aircraft; and (c) a NASA research programme, 1973–1979, that built a method — full-mission simulation — capable of producing error data without killing anyone, and a reporting system — ASRS — capable of producing incident data without punishing anyone. The 1979 NASA/industry workshop at San Francisco named the resulting problem “resource management,” and the training paradigm that replaced individual proficiency training was born out of that naming.

The load-bearing point for any cross-industry argument: aviation could change its training because it had already built the evidence infrastructure. Most industries in 1979 had neither an independent accident investigator with subpoena power, nor a confidential non-punitive reporting channel, nor a high-fidelity rehearsal environment, nor a regulator willing to grant exemptions from its own training rules so that a carrier could try something new.

1. Training before instruments: aptitude, elimination, and the Gosport turn (1903–1930)

1.1 The initial model: flying is a gift

Earliest instruction was not instruction at all. Pupils were given ground runs in a machine with clipped wings, then sent up alone. The instructor’s role was demonstration, not diagnosis. The implicit model was selective: flying ability was a trait one either had or did not, and the function of a flying school was to discover which pupils had it — cheaply, and by attrition.

This model is visible in the RFC’s pre-1917 syllabus. Contemporary and archival accounts describe training that “focused on flying fundamentals, not combat skills or hazard management,” and “taught pilots to observe instructors rather than maximize hands-on control time,” with no preparation for spins, stalls or engine failures — the very conditions that killed them. [SECONDARY — Historic Croydon Airport, drawing on Smith-Barry's own 1917 pamphlet]

1.2 Smith-Barry and the School of Special Flying

Major (later Lt-Col.) Robert Smith-Barry’s intervention in 1916–17 is the first genuine instructional theory in aviation. His premise inverted the selective model: accidents were not evidence of unfitness but of ignorance, and ignorance is remediable. His reported formulation — “Almost all accidents are the result of ignorance” — makes the pupil’s fear a design target rather than a filter. [SECONDARY]

The Gosport system had four components that are still recognisably the architecture of modern flight instruction:

  1. Dual control with the pupil in command of the aircraft, not watching from the passenger seat.
  2. The Gosport tube — a simple speaking tube giving continuous instructor-to-pupil voice contact in flight. Before it, an instructor had to throttle back and shout, which meant instruction was only possible in a condition (low power, low speed) unlike the one being taught.
  3. Deliberate exposure to the dangerous flight regimes — spins, stalls, engine failure, forced landings — under supervision, rather than avoidance of them.
  4. Instruction of the instructors. The School of Special Flying at Gosport existed to train instructors, not pupils. Smith-Barry’s pamphlet General Methods of Teaching Scout Pilots (1917) was circulated through the RFC.

[SECONDARY] for all four; the pamphlet itself is a primary source I could not obtain in full text.

Analytical weight: Gosport is the moment aviation training stops being selection and becomes instruction. It introduces the two principles the whole subsequent history rests on: practise the emergency before you meet it, and standardise the teacher, not just the test. Both reappear, in vastly more expensive form, in the Link Trainer (1934), in LOFT (1975–79), and in CRM (1981–).

1.3 A widely repeated statistic that I could not source

The claim that “over half of the 14,166 pilots killed in the First World War died during training” appears in numerous popular and semi-scholarly accounts, including the Historic Croydon Airport page above. I could not trace it to any primary or peer-reviewed source. [UNVERIFIED] Searches against RFC/RAF casualty statistics, the RAF Museum, and academic WWI aviation literature returned no original derivation. The figure should not be used as load-bearing evidence. It is worth noting explicitly in the final report as an example of how thinly sourced the pre-1940 training-safety record is — which is itself a finding: before formal accident investigation, aviation had anecdote where it later had data.

2. The birth of instrument flight and the simulator (1929–1945)

2.1 Doolittle, 24 September 1929

On 24 September 1929 at Mitchel Field, Garden City, New York, Lt. James H. Doolittle flew a Consolidated NY-2 “Husky” through a complete takeoff, circuit and landing under a hood, with Lt. Benjamin Kelsey aboard as safety pilot. The flight was conducted for the Daniel Guggenheim Fund’s Full Flight Laboratory. [VERIFIED — IEEE Milestone record, which cites the Guggenheim Fund's 1930 final report and NYT of 25 Sept 1929]

The three instruments that made it possible:

  • Kollsman precision barometric altimeter (Kollsman Instrument Co.) — sensitive enough to fly an approach on.
  • Sperry artificial horizon (Sperry Gyroscope Co.) — an attitude reference independent of the pilot’s vestibular sense.
  • Sperry directional gyro — a heading reference free of magnetic compass turning and acceleration errors.

Plus a Bureau of Standards radio range beacon, a marker beacon, and a vibrating-reed receiver by Radio Frequency Laboratories. [VERIFIED — IEEE Milestone record]

Epistemic significance. The blind flight is usually told as a technology story. It is more useful as an epistemology story. Before 1929, the pilot’s body was the primary sensor and the aircraft’s behaviour was read directly. After 1929, the pilot’s body is a demonstrated unreliable sensor, and the primary evidence about the aircraft’s state is a set of representations on a panel. That single move creates every subsequent problem in this literature: instrument scanning, misreading, mode confusion, automation surprise, and the entire discipline of display design. It also creates the first task in aviation that is purely cognitive — believe the instrument, not yourself — and therefore the first task that can be trained on the ground.

2.2 The 1934 air mail crisis as the forcing event

The Link Trainer was invented before there was a market for it. Edwin Link filed his patent on 12 March 1930; it was granted as US 1,825,462 on 29 September 1931. His Link Flying School opened in Binghamton, New York in 1930, and by 1933 the trainer had been adapted for instrument training. Through the Depression the principal customers were amusement parks. [VERIFIED — ASME Historic Mechanical Engineering Landmark brochure, 2000; corroborated by Binghamton University Libraries' Link exhibit]

In February 1934, following cancellation of the commercial air mail contracts, the US Army Air Corps took over air mail carriage. Army pilots — trained for daylight visual military flying — were pushed into night and weather flying with obsolete equipment during exceptionally severe weather. Fatalities followed; the contracts returned to the airlines by mid-1934. The Smithsonian’s National Air and Space Museum account gives “12 airmail pilots killed in accidents” during the training and operating period. [VERIFIED — NASM, though note that other counts circulate and differ depending on whether training, ferry and operational deaths are pooled]

The Army then bought six Link Trainers. [VERIFIED — ASME brochure; Binghamton exhibit] That purchase is the moment simulation entered military aviation procurement.

2.3 What the simulator changed, epistemically

By the end of the Second World War, “over 10,000 ‘blue box’ trainers were used to train more than 500,000 pilots,” with peak production of 80 trainers per week from Binghamton and Canada. [VERIFIED — ASME brochure]

Three changes matter more than the numbers:

  1. Practice without consequence. For the first time, an error could be completed rather than interrupted. In an aircraft, an instructor must take control before the error matures; in a simulator, the error can be allowed to run to its conclusion and then discussed. This is a change in what can be observed, not merely in what can be practised.
  2. Training for rare events. Frequency of exposure in real operations is set by the base rate of the event. In simulation it is set by the syllabus. Engine failure on takeoff — statistically rare, operationally lethal — becomes a routine training item. All later work on low-probability/high-consequence events depends on this.
  3. The instructor becomes an instrument. The Link’s plotter (“crab”) drew the pupil’s track on paper. Performance stopped being the instructor’s recollection and became a record. This is the first data in flight training, and it is the direct ancestor of Ruffell Smith’s 1976 strip-chart printout of aircraft parameters and of modern LOSA/FOQA.

3. Mass production of pilots and the science of selection (1940–1948)

3.1 The Aviation Psychology Program

John Flanagan was recruited in 1940 to build an aviation psychology programme for the US Army; the effort grew out of the Applied Psychology Panel of the National Defense Research Committee and became the AAF Aviation Psychology Program. By summer 1945 it comprised “about 200 officers, 750 enlisted men, and 500 civilians,” headed by Colonels Flanagan, Frank Geldard, J. P. Guilford and Arthur Melton. Its work was published in 1947–48 as a series of 19 volumes known as the “blue books.” [VERIFIED — Roscoe, *The Adolescence of Engineering Psychology*, HFES, pp. 2–3]

The programme’s summary volume is Flanagan, J. C. (Ed.) (1948), The Aviation Psychology Program in the Army Air Forces, AAF Aviation Psychology Program Research Reports No. 1, US GPO, xii + 316 pp. [VERIFIED — bibliographic record via *Psychometrika* review]

3.2 The unrestricted-sample experiment: the strongest evidence in the whole selection literature

In 1943 the AAF, needing throughput, waived the classification test score requirement and admitted a cohort on a first-come basis. The results, as reported in a European Association for Aviation Psychology review drawing on DuBois (1947), Flanagan (1948), Melton (1947) and Guilford & Lacey (1947):

  • 23% of the unselected cohort graduated from advanced training.
  • 63.4% of subsequent (selected) cohorts graduated.
  • Of the 150 cadets admitted with the lowest test scores, none graduated.
  • Cadets in the lowest score third were three times as likely to have accidents as those in the highest third; all four fatalities in the analysis occurred among the lowest-scoring candidates.

[SECONDARY — EAAP, *Selection in Aviation*, citing the AAF blue books. I could not open the blue-book volumes directly; DTIC was unavailable during this research window. These figures should be re-verified against DuBois (1947) before publication.]

3.3 What the washout rate reveals about the model of skill

A washout rate is a confession. It says: we cannot reliably make a pilot out of an arbitrary person, so we will make many attempts and keep the ones that work. An elimination rate of ~37% among selected cadets, and ~77% among unselected ones, tells you that the AAF’s theory of skill was fundamentally dispositional: ability is a stable property of the person, resident before training, measurable in advance, and training’s job is to actualise it rather than create it.

Everything about the institution follows. If ability is dispositional, then:

  • The right lever is selection, and the right science is psychometrics.
  • The right unit of analysis is the individual, not the crew.
  • The right measure of a pilot is a check — a standardised sample of individual performance against a criterion.
  • Failure is diagnostic of the person, not of the system that produced the person.

This model was not stupid. Against the 1940s problem — build 200,000 pilots from a civilian population, fast — it was correct, and it worked. It is also the model that survived, essentially unmodified, into the jet age, where it stopped working. The check-ride architecture of 14 CFR Part 121 (proficiency checks, line checks, Appendix F manoeuvre requirements) is a 1940s dispositional-individual artefact operating in a 1970s multi-crew system. That mismatch is the whole subject of section 8.

4. The birth of aviation human factors (1943–1951)

4.1 Chapanis and the B-17/P-47 gear-flap confusion

The best-sourced account is Roscoe’s:

In 1943, Lt. Alphonse Chapanis was asked why “pilots and copilots of P-47s, B-17s, and B-25s frequently retracted the wheels instead of the flaps after landing.” He “immediately noticed that the side-by-side wheel and flap controls — in most cases identical toggle switches or nearly identical levers — could easily be confused,” and that “the corresponding controls on the C-47 were not adjacent and their methods of actuation were quite different; hence C-47 copilots never pulled up the wheels after landing.” Chapanis “realized that the so-called pilot errors were really cockpit design errors.” The wartime fix attached “a small, rubber-tired wheel … to the end of the wheel control and a small wedge-shaped end to the flap control.” The “mnemonically shape-coded wheel and flap controls were standardized worldwide.”

[VERIFIED — Roscoe, *The Adolescence of Engineering Psychology*, HFES monograph, pp. 2–3]

Two cautions the final report should carry:

  • The “400 B-17s crashed” figure is not sourced. It circulates widely in design-popularisation writing (Inc., Medium, UX trade press). I found no primary source. [UNVERIFIED]
  • The phrase “designer error” is a modern gloss. Roscoe’s wording is “cockpit design errors.” Chapanis’s own first-person account appears in The Chapanis Chronicles (1999), which I could not obtain. Attributing a specific coinage to Chapanis in 1943 is not supportable from what I could read. [UNVERIFIED]

The analytical content survives the sourcing problems intact, and it is the single most important conceptual move in this memo: Chapanis’s inference is a controlled comparison. Same pilots, same stress, same phase of flight; different control geometry; different error rate. That is a natural experiment, and it licenses a causal claim about design rather than about people. Everything the field later does — Fitts & Jones, Fitts 1951, Wiener & Curry, CRM — is downstream of the discovery that you can design out an error class.

4.2 The Aero Medical Laboratory Psychology Branch

“In August 1945, with the war about to end, the AAF Aero Medical Laboratory at Wright Field near Dayton established a Psychology Branch.” Under Lt. Col. Paul Fitts, it included “21 officers, 25 enlisted men, and 10 civilians that first year.” Named staff include Captains Richard E. Jones and H. Richard Van Saun, Majors Walter F. Grether, Judson Brown, Launor Carter and Albert Johnson, and Lieutenants including Robert Gagné and Julien Christensen. Grether succeeded Fitts as technical director in 1949. [VERIFIED — Roscoe, HFES monograph, p. 3]

This matters for citation hygiene: Fitts and Jones were serving officers at the Aero Medical Laboratory, Air Materiel Command, Wright Field/Wright-Patterson, and their 1947 outputs are memorandum reports, not journal articles. Almost every mis-citation in the literature comes from treating them as journal papers.

4.3 The two 1947 Fitts & Jones reports — getting the bibliography right

Report 1 (controls): Fitts, P. M., & Jones, R. E. (1947). Analysis of factors contributing to 460 “pilot-error” experiences in operating aircraft controls. Memorandum Report TSEAA-694-12. Dayton, OH: Aero Medical Laboratory, Air Materiel Command, Wright-Patterson Air Force Base.

  • Title and authorship [VERIFIED — Google Books catalogue record; issuing body given there as "Army Air Forces Headquarters, Air Materiel Command, Engineering Division"; 64 pp.]
  • Report number TSEAA-694-12 [SECONDARY]
  • Method: 460 accounts of errors in operating controls, gathered by interview and written report from AAF pilots — a critical-incident collection. [VERIFIED — Google Books description: "accounts of 460 errors made in operating controls"]
  • Reported finding: “a great many accidents result directly from the manner in which equipment is designed and where it is placed in the cockpit”; “Practically all pilots of present day AAF aircraft, regardless of experience or skill, report that they sometimes make errors.” [SECONDARY — quoted in Shorrock, Humanistic Systems, 2018]
  • Error classes: substitution (operating the wrong control), adjustment errors, forgetting, reversal/direction errors, unintentional activation, inaccessibility. [SECONDARY]

Report 2 (instruments): Fitts, P. M., & Jones, R. E. (1947). Psychological aspects of instrument display. I: Analysis of 270 “pilot-error” experiences in reading and interpreting aircraft instruments. Memorandum Report TSEAA-694-12A. Dayton, OH: Aero Medical Laboratory, Air Materiel Command, Wright-Patterson Air Force Base. [SECONDARY — report number and exact subtitle from Shaver's reference list; I could not open the report. DTIC's full-text service (ADA800143) returned HTTP 403 and then a maintenance page throughout this research window.]

Explicit gap: I could not verify the error-category percentages for either report (e.g. the frequently cited claim that misreading of the three-pointer altimeter dominated the instrument study). Do not use percentage figures for these reports unless someone reads the originals. The commonly cited “40 years later” replication — Flach, J. M., Larish, J. F., & Weinstein, L. F. (1989), Fitts and Jones’ analysis of pilot error: 40 years later, Proceedings of the Fifth International Symposium on Aviation Psychology, 2, 564–569 [VERIFIED — Wright State CORE Scholar record] — reports that the control-error taxonomy generalises across aviation, driving and computer use, whereas the instrument-reading taxonomy does not. That asymmetry is itself worth citing: control errors are about the body’s relation to a device and generalise; display errors are about a specific representational scheme and do not.

4.4 Fitts 1951 and the MABA-MABA list

Fitts, P. M. (Ed.) (1951). Human engineering for an effective air-navigation and traffic-control system. Washington, DC: National Research Council, Division of Anthropology and Psychology, Committee on Aviation Psychology. xii + 84 pp. Prepared with the Ohio State University Research Foundation; related to the US Air Navigation Development Board. [VERIFIED — HathiTrust catalogue record 102215535]

Points of accuracy that are routinely botched:

  • Fitts is the editor of a committee report, not the sole author of a paper. Co-contributors include M. S. Viteles. [VERIFIED — HathiTrust; Semantic Scholar record lists "Fitts, Viteles"]
  • It is an air traffic control report. The famous function-allocation list is an incidental appendix-level device inside a study of the ATC system, not a standalone theory of automation.
  • Fitts did not call it “MABA-MABA.” Dekker & Woods (2002) describe MABA-MABA lists as a genre that “have appeared over the decades in various guises,” of which Fitts (1951) was among the earliest, and do not attribute the acronym to any individual. [VERIFIED — Dekker & Woods full text] I could not establish who coined it. [UNVERIFIED]
  • I could not obtain the eleven list items from the original 1951 text. The reproductions in circulation are paraphrases of paraphrases. [UNVERIFIED — flag as a gap; the primary is in HathiTrust and worth ordering.]

The substantive critique, which the final report should carry because it anticipates the whole automation argument:

Function allocation lists assume “fixed human and machine strengths and weaknesses,” whereas introducing automation “creates new human strengths and weaknesses — often in unanticipated ways.” The result is an “uncooperative system architecture” built on a “‘you do this, I do that’ barter.”

[VERIFIED — Dekker, S. W. A., & Woods, D. D. (2002). MABA-MABA or abracadabra? Progress on human–automation co-ordination. *Cognition, Technology & Work*, 4(4), 240–244.]

5. The jet age: crew complement, the flight engineer, and the authority gradient (1952–1975)

5.1 Crew complement was settled by labour law, not by human factors

Early long-range transports carried five: two pilots, flight engineer, navigator, radio operator. The radio operator disappeared with improved commercial radio; the navigator disappeared with the VOR network in the 1950s and inertial navigation in the late 1960s. The flight engineer’s original raison d’être was the piston engine — “turbo-superchargers and more demanding cooling systems with even more need for in-flight adjustment.” The jet engine “did not require any adjustment in-flight.” [VERIFIED — Fadden, Morton, Taylor & Lindberg, "Evolution of the 2-Person Crew Jet Transport Flight Deck," IEEE ETHW First-Hand account]

That left the third seat’s justification unclear at precisely the moment the jets arrived. The resulting dispute was so severe that President Kennedy intervened. Executive Order 10921, 21 February 1961, established a three-member commission (Nathan Feinsinger, chair; Richard Lester; J. Keith Mann) to inquire into the controversy between Pan American, American, TWA, Eastern, National and Flying Tigers on one side and the Flight Engineers International Association and the Air Line Pilots Association on the other, “regarding the performance of the flight engineer’s function, the job security of employees performing such function, and related representation rights.” [VERIFIED — Executive Order 10921, American Presidency Project]

Analytical point: the composition of the flight deck — the primary determinant of who can catch whose errors — was decided in the early 1960s by a labour arbitration, not by any evidence about crew performance. There was no such evidence to have. It did not exist until Ruffell Smith produced it in 1976. This is the sharpest single illustration of the memo’s third analytical question: where did the field’s evidence come from before formal data programmes existed? Answer: for crew complement, nowhere. It came from bargaining.

The question was finally closed on evidentiary grounds only in 1981, when Secretary of Transportation Drew Lewis’s task force under former FAA Administrator John McLucas concluded that three crew members are not safer than two. [SECONDARY — ETHW First-Hand account; the President's Task Force on Aircraft Crew Complement report (1981) itself was not retrieved and should be obtained.]

5.2 The captain as authority, and the trans-cockpit authority gradient

The doctrine of captain’s authority is ancient maritime law imported into aviation, and it is not in itself the problem. The problem is its gradient — how steeply authority falls off across the flight deck, and what that does to information flow.

The term of art is Elwyn Edwards’s. The earliest citation I could verify is:

Edwards, E. (1975). Stress and the airline pilot. Paper presented to the BALPA Technical Symposium, University of Technology, Loughborough, October 1975. [VERIFIED as a citation — reference list of *Indian Journal of Aerospace Medicine* 52(2), which uses it as the source for "there is an optimum 'trans-cockpit authority gradient' to allow an effective interface between pilots on a flight deck." I did not read Edwards's paper itself.]

Edwards is the same author who produced the SHEL model (1972, elaborated 1975); Frank Hawkins subsequently built KLM’s human factors training on Edwards’s SHEL model and TAG concept. [SECONDARY — Helmreich & Foushee, "Why CRM?", ch. 1 of *Crew Resource Management*, Elsevier, 2010]

The concept is explicitly two-tailed, and the final report should preserve that, because the popular telling only remembers one tail:

  • Too steep: a dominant senior captain and an unassertive junior officer; the junior withholds concerns out of deference. (Tenerife; United 173.)
  • Too flat: two equally qualified individuals, with the result that neither takes command and errors go undetected. [VERIFIED — *Indian Journal of Aerospace Medicine* 52(2)]

6. The accidents that forced a new paradigm (1972–1978)

I treat these as evidence, and cite the official reports.

6.1 Eastern Air Lines 401, Everglades, 29 December 1972

NTSB-AAR-73-14, adopted 14 June 1973. Lockheed L-1011, N310EA. 176 aboard; 99 fatal at the scene, two further deaths from injuries. Probable cause, verbatim:

“The failure of the flightcrew to monitor the flight instruments during the final 4 minutes of flight, and to detect an unexpected descent soon enough to prevent impact with the ground. Preoccupation with a malfunction of the nose landing gear position indicating system distracted the crew’s attention from the instruments and allowed the descent to go unnoticed.”

[VERIFIED — OCR of the NTSB report's technical title page]

The design substrate matters as much as the crew behaviour, and it is the clearest early instance of automation surprise:

  • The L-1011 autopilot reverted from altitude hold to pitch Control Wheel Steering on an applied control-column force — 15 lb on autopilot A, 20 lb on autopilot B — while “the engagement lever will remain in the originally selected position,” so “it is possible to disengage altitude hold without an accompanying alert to the flight crew.”
  • The altitude alert was a single C-chord plus a flashing amber light; on Eastern’s configuration the light was inhibited below 2,500 ft radar altitude, so “the only altitude alert system warning to the flight crew that the airplane was descending was the single C-chord.”

[VERIFIED — FAA *Lessons Learned from Transport Airplane Accidents*, N310EA entry]

NASA’s own reading of the accident, in Lauber’s 1979 paper, quotes the NTSB findings including: “the aircraft was flown to a safe altitude, and the autopilot was engaged to reduce workload, but positive delegation of aircraft control was not accomplished”; and “the captain failed to assure that a pilot was monitoring the progress of the aircraft at all times.” [VERIFIED — Lauber, in NASA CP-2120, pp. 5–6]

6.2 TWA 514, Berryville, Virginia, 1 December 1974

NTSB-AAR-75-16, report dated 26 November 1975. Boeing 727-231, N54328. 92 killed. Probable cause, verbatim:

“The crew’s decision to descend to 1,800 feet before the aircraft had reached the approach segment where that minimum altitude applied. The crew’s decision to descend was a result of inadequacies and lack of clarity in the air traffic control procedures which led to a misunderstanding on the part of the pilots and of the controllers regarding each other’s responsibilities during operations in terminal areas under instrument meteorological conditions.”

Contributing: FAA’s failure to resolve known terminology confusion; approach clearance issued 44 miles out on an unpublished route without defined minimum altitudes; inadequate altitude depiction on the profile view of the approach chart. [VERIFIED — OCR of the NTSB technical title page]

TWA 514 is the founding document of the reporting infrastructure, and the relevant passage is section 1.17.8, “Unsafe-Condition Reporting and Investigating,” which I quote at length because it is decisive:

“In January 1974, an air carrier in the United States initiated a Flight Safety Awareness Program … Under this program, an individual could make a report without identifying himself or his fellow crewmembers. The pilots were assured that the carrier would not take any punitive action as a result of information procured through this program …

In October 1974, the carrier received a report under this program. A crew reported that they were approaching Dulles and after passing Front Royal at 6,500 feet, they were issued a clearance to descend to 4,000 feet … they were cleared for a VOR approach to runway 12. After the captain reviewed the chart for the latter approach, he descended to 1,800 feet … and landed without incident. After landing, the crew reviewed the approach and decided that they had descended to 1,800 feet about 25 nmi from the VOR and were at that altitude before they reached the Round Hill intersection …

As a result of this investigation, the carrier believed it was not necessary to make any recommendations to the FAA or to change the carrier’s procedures. However, they did publish a notice to all flightcrews …

Except for regulatory reporting requirements that aircrews notify the nearest ground station when an irregularity is noted in a navigational facility or ground facility, the FAA had no formal system for pilots or controllers to report unsafe conditions involving instrument flight procedures in the terminal area. Witnesses testified that reports of unsafe conditions were not furnished to the FAA or to the carriers because the individuals were afraid of punitive action. These witnesses recommended that the FAA establish a system to enable pilots and controllers to report operational hazards with immunity provided for the person making the report.”

[VERIFIED — OCR of NTSB-AAR-75-16, §1.17.8, pp. 22–24]

Six weeks before 92 people died, another carrier’s crew had made the identical error, survived it, reported it internally, and the information had stopped at the company boundary. The NTSB report notes as a post-accident FAA action that “the FAA has established an incident reporting system which is intended to identify unsafe operating conditions in order that they can be corrected before an accident occurs.” [VERIFIED — NTSB-AAR-75-16, §3]

6.3 Tenerife, Los Rodeos, 27 March 1977

KLM Boeing 747 PH-BUF and Pan Am Boeing 747 N736PA; 583 killed. The Spanish Subsecretaría de Aviación Civil report (issued October 1978) attributes the fundamental cause to the KLM captain taking off without clearance, ignoring the tower’s “stand by,” continuing despite the Pan Am transmission that it was still on the runway, and — critically for the training argument — responding “emphatically in the affirmative” when his flight engineer asked whether the Pan Am aircraft had cleared. [VERIFIED — "Secretary of Aviation Report on Tenerife Crash," hosted by FAA]

Two contributing mechanisms named: non-standard phraseology (“we are now at take-off,” which the controller did not read as a takeoff roll), and simultaneous transmissions producing a heterodyne that masked both the tower’s “stand by for take-off” and Pan Am’s “we are still taxiing down the runway.” [VERIFIED — same source]

Tenerife’s contribution to the training argument is not the gradient story alone. It is that the KLM captain was KLM’s chief of flight training — the most authoritative technical figure in the airline — and that his technical excellence was precisely what made the flight engineer’s challenge ineffective. Expertise is a component of the gradient, not a substitute for the challenge. This is the fact that makes “more training in flying skill” a non-answer.

6.4 United Airlines 173, Portland, 28 December 1978 — the pivot

NTSB-AAR-79-7, adopted 7 June 1979. McDonnell Douglas DC-8-61, N8082U. (Note: Helmreich’s widely cited RAeS paper mis-states the type as DC-8-54; the NTSB title page says DC-8-61.) [VERIFIED — NTSB title page vs. Helmreich 2003 reference list]

Probable cause, verbatim:

“The failure of the captain to monitor properly the aircraft’s fuel state and to properly respond to the low fuel state and the crewmember’s advisories regarding fuel state. This resulted in fuel exhaustion to all engines. His inattention resulted from preoccupation with a landing gear malfunction and preparations for a possible landing emergency.

Contributing to the accident was the failure of the other two flight crewmembers either to fully comprehend the criticality of the fuel state or to successfully communicate their concern to the captain.”

[VERIFIED — NTSB-AAR-79-7, §3.2]

The analysis section is where the paradigm shift is actually written down, and it deserves quoting in the final report:

“The Safety Board believes that this accident exemplifies a recurring problem — a breakdown in cockpit management and teamwork during a situation involving malfunctions of aircraft systems in flight. To combat this problem, responsibilities must be divided among members of the flightcrew while a malfunction is being resolved. In this case, apparently no one was specifically delegated the responsibility of monitoring fuel state.

Although the captain is in command and responsible for the performance of his crew, the actions or inactions of the other two flight crewmembers must be analyzed. Admittedly, the stature of a captain and his management style may exert subtle pressure on his crew to conform to his way of thinking. It may hinder interaction and adequate monitoring and force another crewmember to yield his right to express an opinion.

The first officer’s main responsibility is to monitor the captain. In particular, he provides feedback for the captain. If the captain infers from the first officer’s actions or inactions that his judgment is correct, the captain could receive reinforcement for an error or poor judgment …

The first officer’s and the flight engineer’s inputs on the flight deck are important because they provide redundancy. The Safety Board believes that, in training of all airline cockpit and cabin crewmembers, assertiveness training should be a part of the standard curricula, including the need for individual initiative and effective expression of concern.”

[VERIFIED — NTSB-AAR-79-7, §2 Analysis]

And the recommendation itself, verbatim from the report (OCR reconstructed; bracketed characters were degraded in the scan):

“Issue an operations bulletin to all air carrier operations inspectors directing them to u[rge] their assigned operators to ensure that their flightcrews are indoctrinated in principles of flight deck resource management, with particular emphasis on the merits of participative management for captains and assertiveness training for other cockpit crewmembers. (Class II, Priority Action) (A-79-4[7])”

[VERIFIED for the text; the recommendation number is rendered ambiguously by the OCR of the scanned report and is cited in the secondary literature as A-79-47 — treat the number as [SECONDARY] and confirm against NTSB's recommendation database before publication.]

Three things make this the pivot, and all three are about language:

  1. It names “flight deck resource management” as a training object — three weeks before the NASA workshop that is usually credited with coining the term.
  2. It prescribes two distinct interventions for two distinct roles: participative management for captains, assertiveness for everyone else. The gradient is being addressed from both ends.
  3. It routes the intervention through the FAA’s operations inspectors — i.e. it makes a behavioural, non-technical training topic into a matter of regulatory oversight. That is the mechanism by which a psychological finding becomes a required curriculum.

7. Building the evidence base: interviews, ASRS, and full-mission simulation (1973–1979)

7.1 The pilot interview study, from 1973

NASA Ames began its human-factors-in-aviation-safety programme in 1973 with structured, confidential interviews with airline crewmembers, run with the cooperation of both management and unions. Lauber’s summary of what came back:

“Generally, those pilots who mentioned training during the interviews expressed satisfaction with the training they receive in the technical aspects of flying and in flying skills. The difficulty related more to issues such as how to be a more effective leader, and how to achieve more effective crew coordination and improved communication within the cockpit. One new captain stated the problem as follows: ‘My company trains pilots very well, but not captains — command training is needed.’ A flight engineer felt strongly that ‘Pilots, particularly captains, ought to be given specific training in human behavior and human relationships … there is too much emphasis on the technical side of flying, and too little upon these “softer” issues.’”

[VERIFIED — Lauber, "Resource Management on the Flight Deck: Background and Statement of the Problem," NASA CP-2120, p. 3]

This is the whole thesis of the report in one quotation, stated by pilots, six years before the workshop. The industry’s own practitioners had diagnosed the gap correctly; what was missing was a method for demonstrating it.

7.2 The Aviation Safety Reporting System, 15 April 1976

The FAA established a voluntary reporting programme in 1975 but concluded that its own regulatory role would suppress participation, and “asked NASA to act as the highly respected, independent third party” — an “honest broker.” ASRS began operation on 15 April 1976, under a Memorandum of Agreement between the FAA and NASA. [VERIFIED — NASA ASRS Publication 60, "ASRS: The Case for Confidential Incident Reporting Systems"]

The design principles, which are the transferable part:

  • Third-party custody. The receiving, processing and analysis of raw data is done by NASA, not the regulator. [VERIFIED — ASRS immunity page]
  • De-identification. “All information that might assist in or establish the ID of persons filing ASRS reports and parties named in those reports will be deleted,” with narrow exceptions for criminal activity and accidents. An identification strip is returned to the reporter as receipt. [VERIFIED — ASRS immunity page]
  • Conditional enforcement immunity, under FAA Advisory Circular 00-46 (current revision 00-46F; earlier 00-46E dated 16 December 2011), conditional on the violation being inadvertent and non-deliberate, not criminal, not involving an accident or lack of qualification, no prior enforcement finding in five years, and a report filed within ten days. [VERIFIED — ASRS immunity page]

By April 2001 ASRS had processed its 500,000th report; by 2002, over 558,000. [VERIFIED — ASRS Pub. 60]

Why this mattered for training. By June 1979, NASA could search the ASRS database — then holding “nearly 7,000 reports” — for jet transport operations and recover 670 incident reports relevant to resource management. [VERIFIED — Lauber, CP-2120, p. 7] For the first time a training argument could be made from a population of events rather than from a handful of catastrophes. ASRS is what converted the resource-management problem from a coroner’s inference into a base rate.

7.3 Ruffell Smith 1979 — the load-bearing study

Citation (get this right): Ruffell Smith, H. P. (1979). A simulator study of the interaction of pilot workload with errors, vigilance, and decisions. NASA Technical Memorandum 78482 (also numbered A-7354). Moffett Field, CA: NASA Ames Research Center, January 1979. Report signed off at Ames 24 August 1978. [VERIFIED — full text read: https://ntrs.nasa.gov/api/citations/19790006598/downloads/19790006598.pdf]

Design, from the report itself:

  • Proposed to NASA Ames by the author in 1974; supported by a National Research Council associateship; preparations from July 1975; experiment conducted early in 1976. (The four-year gap between execution and publication explains why some citations date it 1976 or 1978.)
  • Twenty fully qualified three-man civil airline crews.
  • Simulator: Singer-Link, six-degrees-of-freedom motion platform, long-haul flight deck, with visual attachment permitting transition from instrument to visual on approach, plus stations for pilot and engineer instructor/operators. The report deliberately says only “a contemporary wide-body transport aircraft.” Internal evidence identifies it as a Boeing 747: gross weight figures of 564,000 lb (the 747-100 maximum landing weight) and 647,000 lb appear in the error narratives. Lauber, writing in CP-2120, states plainly that it used “a B-747 training simulator.” [VERIFIED — report text + Lauber, CP-2120, p. 4]
  • Scenario: a charter, Dulles → JFK (fuel stop) → Heathrow. Sector 1 low workload, mean 47 min. Sector 2 high workload, mean 97 min, including heavy weight, marginal runway, a complex departure, a No. 2 engine oil-filter clog requiring shutdown, No. 3 hydraulic system depletion (which removed the remaining autopilot), fuel dumping, diversion, and a demanding approach. Realism aids included twelve tapes of genuine recorded ATC party-line traffic per sector, cabin-attendant interruptions on the interphone, and a simulated dispatcher.
  • Measures: observer commentary on strip chart, crew paperwork, continuous ECG on all three crew members, all flight-deck and air-ground communications, and a continuous printout of aircraft parameters.

Findings, verbatim from the report’s Summary:

“The results showed that the number of errors was very variable among crews but the mean increased in the higher workload case. The increase in errors was not related to rise in heart rate but was associated with vigilance times as well as the days since the last flight.

The recorded data also made it possible to investigate decision time and decision order. These also varied among crews and seemed related to the ability of captains to manage the resources available to them on the flight deck.

Error rates and heart rates were essentially the same as those found in actual flight operations, indicating the quality of the simulation.”

[VERIFIED]

Errors were about 2.5 times more numerous in the high-workload sector, concentrated in tactical decisions, systems operation and navigation. [VERIFIED — Results, p. 15]

On captaincy — the passage that gave the field its vocabulary:

“For the first time this experiment provided opportunities to observe 18 different captains as they responded to the demands of the same abnormal flight conditions. There seemed to be large variations in respect to leadership, resource management and decisionmaking … In some of the runs leadership quality seemed lacking and on occasion P2 appeared to have usurped the role of P1 … When P1s were aware of the situation and delegated their P2s to controlling the aircraft in its flight path, immediate benefits were evident.”

[VERIFIED — Ruffell Smith, pp. 28–29]

Conclusions:

“Many of the problems, however, relate to the management of human and mechanical resources. The variability between crews in reacting to the same problems suggests that those who perform less well might be helped by special training. Consequently, it is recommended that … special training in resource management and captaincy be developed and validated. Such training should include the use of full mission simulation of scenarios that are representative of actual situations.”

[VERIFIED — Ruffell Smith, pp. 34–35]

Two citation cautions.

  • The report says twenty crews but “18 different captains.” Helmreich & Foushee’s widely reproduced summary says “eighteen airline crews.” Both numbers are in circulation; the report supports 20 crews / 18 captains. [VERIFIED discrepancy]
  • The often-repeated claim that “crew coordination, not technical skill, caused most errors” is Helmreich & Foushee’s gloss (“most problems and errors were induced by breakdowns in crew coordination rather than by deficits in technical knowledge and skills”). Ruffell Smith’s own conclusion is more careful: he attributes difficulties to flight-deck and instrument design, to documents and charts, and to resource management. Do not attribute the stronger claim to the 1979 report. [VERIFIED — compare Ruffell Smith conclusions with Helmreich & Foushee ch. 1]

Why this study is load-bearing. It is the first time in aviation that (i) a representative sample of ordinary line crews (ii) faced an identical, non-trivial, realistic problem (iii) with complete instrumentation of behaviour and physiology (iv) at no risk. It converted “crew coordination” from a folk concept into a measured variable with observable between-crew variance. Variance is the thing: if all crews had performed identically, there would be nothing to train. Twenty crews, one scenario, wildly different outcomes — that is a training warrant.

8. The 1979 NASA/industry workshop and the naming of the problem

Citation: Cooper, G. E., White, M. D., & Lauber, J. K. (Eds.) (1980). Resource Management on the Flight Deck: Proceedings of a NASA/Industry Workshop. NASA Conference Publication 2120. Moffett Field, CA: NASA Ames Research Center. Workshop held at the Jack Tar Hotel, San Francisco, California, 26–28 June 1979; proceedings published March 1980. Cochairs: John K. Lauber (NASA Ames) and Capt. A. A. Frink (Pan American). Organised by G. E. Cooper Associates for the Aviation Safety Research Office, Ames. [VERIFIED — full text read: https://ntrs.nasa.gov/api/citations/19800013796/downloads/19800013796.pdf]

8.1 What it actually was

Not a conference. Four sessions: (1) non-industry keynotes on the human factors of the problem — Lauber, Robert Helmreich (Texas), Lee Bolman (Harvard); (2) airline descriptions of what they were already doing — United, British Airways, American, Swissair, North Central, plus Berton Beach of Eastern on Line-Oriented Flight Training; (3) structured working groups organised by training phase (selection/initial, recurrent/line, transition/upgrade) charged with producing conclusions and recommendations; (4) a summary by Charles Billings. Participants were “senior officers of major airlines who are responsible for aircrew training; representatives of cognizant government agencies; and specialists in human factors work as it applies to aircrew operations.” Some papers are verbatim transcript. [VERIFIED — Preface and Contents, CP-2120]

This structure is the finding. It was not a dissemination event; it was a design event, with the customers of the eventual product in the room doing the work.

8.2 How it named the problem

Lauber’s opening thesis:

“One of the principal causes of incidents and accidents in civil jet transport operations is the lack of effective management of available resources by the flight-deck crew. It is further argued that present aircrew training programs could be augmented to improve flight-deck management.”

[VERIFIED — CP-2120, p. 3]

The evidence he assembled was tripartite and is exactly the three-source structure of the whole field: interviews (§7.1), accidents — a NASA review that “identified more than 60 accidents in the period 1968–1976 in which resource management problems played a significant role,” with a full list at Table 1 — and incidents (670 ASRS reports). [VERIFIED — CP-2120, pp. 5–7]

From these he derived the operational definition, as a list of the most frequently observed problems:

“Preoccupation with minor mechanical problems; Inadequate leadership; Failure to delegate tasks and assign responsibilities; Failure to set priorities; Inadequate monitoring; Failure to utilize available data; Failure to communicate intent and plans.”

[VERIFIED — CP-2120, p. 7]

That list is the first curriculum. Every subsequent CRM syllabus is a rearrangement of it.

Terminology note, precisely. The workshop’s own title and Lauber’s framing are “resource management on the flight deck” / “flight-deck resource management.” The phrase “cockpit resource management” appears exactly once in the 306-page proceedings, inside a working-group report — i.e. it was in the room but not yet the label. [VERIFIED — full-text search of CP-2120] The NTSB had already published “flight deck resource management” in its United 173 report adopted 7 June 1979, nineteen days before the workshop opened. [VERIFIED] The common assertion that the workshop “coined CRM” should therefore be softened to: the workshop consolidated and institutionalised a term already in circulation.

8.3 LOFT was already running, and the regulation was the obstacle

Capt. Berton E. Beach of Eastern told the workshop:

“We recognize the need to shift from training in manipulative skills to something closer to management skills. Line-Oriented Flight Training (LOFT) is not a new idea. We used a similar format at Eastern in the late 1950’s on our DC-8 and Boeing 720 series aircraft. At that time, the simulators available had no motion, nor visual capabilities …

Our first effort to implement LOFT was a scenario we developed in 1975, wherein the crew took a three-leg, four-hour flight that satisfied all requirements of Appendix F, except for steep turns and approaches to stall … Shortly after this time, Eastern applied for, but was not granted, permission to operate under an exemption from Appendix F …

Our current format was developed after guidelines for exemption were published, and we found that we were not required to conform to Appendix F at all except that we must maintain our landing certification to Category II or Category IIIa minimums.”

[VERIFIED — CP-2120, pp. 107–108]

This is the single most transferable institutional fact in the memo. The technical idea (rehearse the whole job in real time, unbroken, with the crew’s own errors allowed to develop) existed at Eastern in the late 1950s. It was blocked for twenty years by two things: simulator fidelity, and a regulation — Part 121 Appendix F — that specified manoeuvres to be demonstrated, which is a manoeuvre-based, individual, dispositional conception of competence. LOFT became possible when the regulator published exemption guidelines — that is, when the rule acquired a legal path around itself. The technology mattered; the exemption mattered more.

8.4 Billings’s summary: what the workshop concluded it did not know

Charles Billings’s closing summary is unusually candid and worth citing because it shows the field understood its own limits from day one:

“The goal here … is to assist aircrewmembers to recognize that discordant situations can exist, to assist them to create an environment that maximizes the likelihood that they will be recognized, and to assist them to create an approach that will permit them to diagnose such discord, to find the reasons why everything doesn’t quite add up.”

“One of the beauties of LOFT is that the captain and the crew get to dig their own hole. But … one of the most serious problems with LOFT, for those people whose resources for dealing with those kinds of holes are marginal, is going to be helping them out of the hole, helping them to recognize why they fell in the hole in the first place, and helping them to fill in the hole. And all three of those have got to be done. This … places a very considerable burden on the LOFT instructors and on their training.”

“No one approach to this business is going to work in all carriers … We’re talking about the bottom 15 or 20 percent. It is important to recognize the difference between a psychological problem and a proficiency problem because the treatment of the two may be very different.”

[VERIFIED — CP-2120, pp. 194–196]

Billings identified in June 1979 the three failure modes that would in fact plague CRM for the next twenty years: instructor capability as the binding constraint; the debrief, not the scenario, as the locus of learning; and the impossibility of a single programme for all carriers and all individuals.

9. Answers to the three analytical questions

9.1 What model of “the pilot” was implicit before 1980, and what replaced it?

Before: the pilot as a selected individual with trainable manual and procedural skill.

Its markers, all verifiable in the record above:

  • Selection science as the primary lever (AAF classification battery; 23% vs 63.4% graduation).
  • Elimination as the primary quality mechanism (washout).
  • The individual proficiency check against a manoeuvre list as the primary certification (Part 121 Appendix F).
  • Authority allocated by seat and seniority rather than by information (EO 10921 settled crew composition by arbitration; the captain’s authority gradient was unexamined until Edwards named it in 1975).
  • “Pilot error” as an explanatory terminus rather than a starting point.

After: the pilot as one node in a crew–machine–procedure system whose output is a joint product.

Its markers:

  • The unit of analysis moves from person to crew (Ruffell Smith’s between-crew variance; UA173’s “no one was specifically delegated the responsibility of monitoring fuel state”).
  • The object of training moves from manoeuvres to management (Beach: “shift from training in manipulative skills to something closer to management skills”).
  • The criterion moves from demonstrated proficiency to observed process (LOFT: unbroken real-time scenario, crew errors allowed to develop, debrief as the intervention).
  • Error is reclassified as a system output with design and organisational causes (Chapanis 1943; Fitts & Jones 1947; the L-1011’s silent altitude-hold reversion).
  • Authority becomes an engineered variable with an optimum, not a fixed prerogative (Edwards’s TAG; NTSB’s paired prescription of participative management and assertiveness).

The two models are not sequential in the way textbooks suggest. The design critique (Chapanis, Fitts & Jones) had refuted the individual-blame model by 1947 — thirty years before the training system acted on it. That thirty-year lag is the interesting fact.

9.2 What made aviation able to change its training at all?

Five institutional preconditions existed by 1979 that few other industries had, and the report should present them as a checklist:

  1. An independent, non-regulatory, non-prosecutorial investigator with statutory access. The NTSB could subpoena, publish, name causes, and issue recommendations to the regulator without being the regulator. The UA173 report’s analysis section is a piece of published organisational psychology commissioned by nobody, protected by statute. No other sector in 1979 had an equivalent.
  2. A confidential, non-punitive, third-party-held incident reporting channel. ASRS from 15 April 1976. The critical design choice — that NASA, not the FAA, held the raw data — is what made the base rate obtainable. The TWA 514 report shows exactly what happens without it: a correct diagnosis reaching one company and stopping there.
  3. A high-fidelity, zero-consequence rehearsal environment with instrumentation. Six-degree-of-freedom motion plus a visual system, by 1976. Not a training convenience — a laboratory. Ruffell Smith’s result was only possible because it existed.
  4. A regulator willing to permit deviation from its own training rules. Eastern’s LOFT programme was blocked by Appendix F and enabled by published exemption guidelines. Regulatory flexibility was the rate-limiting step, not regulatory stringency.
  5. A convening body trusted by both management and labour. NASA Ames could run confidential interviews “with the full cooperation of management and union representatives,” and could put airline training chiefs, the FAA and academic psychologists in one room for three days of working groups. In an industry with the labour history that produced EO 10921, a neutral convener was not optional.

To which a sixth, softer condition should be added: the industry’s own practitioners had already made the diagnosis (“My company trains pilots very well, but not captains”). The research programme did not have to persuade the field that a problem existed; it had to supply the evidence and the method that made action defensible.

9.3 Where did the evidence come from before formal data programmes existed?

Four sources, in rough order of appearance, each with a characteristic bias the final report should name:

  1. Accident reports (1926– ). Rich, causally analysed, publicly available — and radically censored by survivorship. They describe only the sequences that ran to completion, in a heavily non-random sample. The NTSB’s UA173 phrase “this accident exemplifies a recurring problem” is a claim the report cannot itself support; it borrows credibility from an accumulated impression across many investigations.
  2. The critical-incident technique (1943– ). Fitts & Jones’s 460 and 270 error accounts, collected by asking pilots to recount errors. Flanagan formalised the method during the same programme. It reaches near-misses that accidents miss, but depends on recall and on the reporter’s willingness to self-incriminate — which is why ASRS’s immunity architecture was necessary before the technique could scale.
  3. Structured confidential interviews (NASA Ames, 1973– ). Union-and-management-sanctioned, non-attributable, qualitative. Excellent for problem identification, useless for rates. This is what produced the diagnosis; it could not have produced the warrant.
  4. Full-mission simulation (1976). The first source that supplies a controlled comparison: identical stimulus, twenty crews, complete instrumentation, no survivorship censoring, no recall bias. Its own bias is volunteerism — Ruffell Smith’s subjects were volunteers, about 30% of available aircrew at their domicile, not a random sample — and the Hawthorne risk of observed performance. But it is the only pre-1980 source capable of estimating variance between crews, which is the quantity that justifies training.

The general lesson for a cross-industry argument: aviation did not wait for an ideal data source. It layered four imperfect ones with non-overlapping biases, and treated their convergence as the warrant. Lauber’s 1979 paper is literally structured that way — interviews, then 60+ accidents, then 670 incidents, then the simulation — and the convergence, not any single source, is what made the case.

10. Gaps, unresolved questions, and unverified claims

Listed separately in the covering note. In brief: the Fitts & Jones originals (error percentages, exact report numbers) could not be read because DTIC full text was unavailable; the Fitts 1951 list items could not be obtained from the primary; the Chapanis “designer error” coinage and the “400 B-17s” figure are unsourced; the WWI “14,166 / over half in training” figure is unsourced; the UA173 recommendation number is degraded in the scanned original; the AAF washout figures are held only via a secondary review of the blue books; and the 1961 Feinsinger Commission’s actual recommendations and the 1981 McLucas task force report were not retrieved.

11. Bibliography

Primary — official reports and technical memoranda

  1. Cooper, G. E., White, M. D., & Lauber, J. K. (Eds.) (1980). Resource Management on the Flight Deck: Proceedings of a NASA/Industry Workshop, San Francisco, June 26–28, 1979. NASA Conference Publication 2120. Moffett Field, CA: NASA Ames Research Center. https://ntrs.nasa.gov/citations/19800013796 (full text: https://ntrs.nasa.gov/api/citations/19800013796/downloads/19800013796.pdf) [VERIFIED]
  2. Ruffell Smith, H. P. (1979). A simulator study of the interaction of pilot workload with errors, vigilance, and decisions. NASA TM-78482 (A-7354). Moffett Field, CA: NASA Ames Research Center. https://ntrs.nasa.gov/citations/19790006598 (full text: https://ntrs.nasa.gov/api/citations/19790006598/downloads/19790006598.pdf) [VERIFIED]
  3. National Transportation Safety Board (1979). Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978. NTSB-AAR-79-7. Adopted 7 June 1979. https://www.ntsb.gov/investigations/AccidentReports/Reports/AAR7907.pdf [VERIFIED]
  4. National Transportation Safety Board (1975). Aircraft Accident Report: Trans World Airlines, Inc., Boeing 727-231, N54328, Berryville, Virginia, December 1, 1974. NTSB-AAR-75-16. Dated 26 November 1975. https://www.ntsb.gov/investigations/AccidentReports/Reports/AAR7516.pdf [VERIFIED]
  5. National Transportation Safety Board (1973). Aircraft Accident Report: Eastern Air Lines, Inc., L-1011, N310EA, Miami, Florida, December 29, 1972. NTSB-AAR-73-14. Adopted 14 June 1973. https://www.ntsb.gov/investigations/AccidentReports/Reports/AAR7314.pdf [VERIFIED]
  6. Subsecretaría de Aviación Civil, Ministerio de Transportes y Comunicaciones (Spain) (1978). Report on the collision between KLM Boeing 747 PH-BUF and Pan Am Boeing 747 N736PA at Los Rodeos Airport, Tenerife, 27 March 1977 [“Secretary of Aviation Report on Tenerife Crash”], October 1978. https://www.faa.gov/sites/faa.gov/files/2022-11/Spanish_Findings_0.pdf [VERIFIED]
  7. Wiener, E. L., & Curry, R. E. (1980). Flight-deck automation: Promises and problems. NASA TM-81206 (A-8210), June 1980. https://ntrs.nasa.gov/citations/19800017542 [VERIFIED — bibliographic record and abstract]
  8. Fitts, P. M. (Ed.) (1951). Human engineering for an effective air-navigation and traffic-control system. Washington, DC: National Research Council, Division of Anthropology and Psychology, Committee on Aviation Psychology; prepared with the Ohio State University Research Foundation. xii + 84 pp. https://catalog.hathitrust.org/Record/102215535 [VERIFIED — catalogue record only; text not read]
  9. Fitts, P. M., & Jones, R. E. (1947). Analysis of factors contributing to 460 “pilot-error” experiences in operating aircraft controls. Memorandum Report TSEAA-694-12. Dayton, OH: Aero Medical Laboratory, Air Materiel Command, Wright-Patterson AFB. 64 pp. Catalogue: https://books.google.com/books/about/Analysis_of_Factors_Contributing_to_460.html?id=XjgrygAACAAJ [VERIFIED title/authors/year; report number [SECONDARY]; text not read]
  10. Fitts, P. M., & Jones, R. E. (1947). Psychological aspects of instrument display. I: Analysis of 270 “pilot-error” experiences in reading and interpreting aircraft instruments. Memorandum Report TSEAA-694-12A. Dayton, OH: Aero Medical Laboratory, Air Materiel Command, Wright-Patterson AFB. [SECONDARY]
  11. Flanagan, J. C. (Ed.) (1948). The Aviation Psychology Program in the Army Air Forces. Army Air Forces Aviation Psychology Program Research Reports, Report No. 1. Washington, DC: US Government Printing Office. xii + 316 pp. Record: https://www.cambridge.org/core/journals/psychometrika/article/abs/john-c-flanaganeditor-the-aviation-psychology-program-in-the-army-air-forces-…/8D0DB5ABB36B6AC581690BD29FE254B3 [VERIFIED — bibliographic record]
  12. Kennedy, J. F. (1961). Executive Order 10921 — Establishing a Commission to Inquire Into a Controversy Between Certain Air Carriers and Certain of Their Employees, 21 February 1961. https://www.presidency.ucsb.edu/documents/executive-order-10921-establishing-commission-inquire-into-controversy-between-certain-air [VERIFIED]
  13. NASA Aviation Safety Reporting System (n.d.). ASRS: The Case for Confidential Incident Reporting Systems. ASRS Publication 60. https://asrs.arc.nasa.gov/docs/rs/60_Case_for_Confidential_Incident_Reporting.pdf [VERIFIED]
  14. NASA Aviation Safety Reporting System (n.d.). Immunity Policies [FAA AC 00-46 series; report handling and de-identification]. https://asrs.arc.nasa.gov/overview/immunity.html [VERIFIED]
  15. Federal Aviation Administration (n.d.). Lessons Learned from Transport Airplane Accidents: Lockheed L-1011, N310EA. https://www.faa.gov/lessons_learned/transport_airplane/accidents/N310EA [VERIFIED]

Secondary and scholarly

  1. Roscoe, S. N. (1997). The Adolescence of Engineering Psychology. Human Factors History Monograph Series, Vol. 1. Santa Monica, CA: Human Factors and Ergonomics Society. https://www.hfes.org/Portals/0/Documents/adolescence.pdf [VERIFIED]
  2. Dekker, S. W. A., & Woods, D. D. (2002). MABA-MABA or abracadabra? Progress on human–automation co-ordination. Cognition, Technology & Work, 4(4), 240–244. DOI: 10.1007/s101110200022. Author copy: https://www.humanfactors.lth.se/fileadmin/lusa/Sidney_Dekker/articles/2003_and_before/MABA_MABA.pdf [VERIFIED]
  3. Helmreich, R. L., & Foushee, H. C. (2010). Why CRM? Empirical and theoretical bases of human factors training. Ch. 1 in B. Kanki, R. Helmreich & J. Anca (Eds.), Crew Resource Management (2nd ed.). Amsterdam: Elsevier. https://booksite.elsevier.com/samplechapters/9780123749468/9780123749468.pdf [VERIFIED]
  4. Helmreich, R. L. (2003). The Evolution of Crew Resource Management Training in Commercial Aviation. Royal Aeronautical Society Centennial Human Factors Group paper, 15 October 2003. https://www.raes-hfg.com/reports/15oct03-Centennial/15oct03-RHelmreich.pdf [VERIFIED — note it mis-states UA173's aircraft as DC-8-54]
  5. Flach, J. M., Larish, J. F., & Weinstein, L. F. (1989). Fitts and Jones’ analysis of pilot error: 40 years later. Proceedings of the Fifth International Symposium on Aviation Psychology, 2, 564–569. https://corescholar.libraries.wright.edu/psychology/372/ [VERIFIED — record and abstract]
  6. Edwards, E. (1975). Stress and the airline pilot. Paper presented to the BALPA Technical Symposium, University of Technology, Loughborough, October 1975. [SECONDARY — cited as the origin of "trans-cockpit authority gradient" in Indian J. Aerosp. Med. 52(2), https://indjaerospacemed.com/content/110/2008/52/2/pdf/IJASM-52-079.pdf]
  7. Fadden, D. M., Morton, P. M., Taylor, R. W., & Lindberg, T. (2014). First-Hand: Evolution of the 2-Person Crew Jet Transport Flight Deck. IEEE Engineering and Technology History Wiki. https://ethw.org/First-Hand:Evolution_of_the_2-Person_Crew_Jet_Transport_Flight_Deck [VERIFIED]
  8. IEEE (n.d.). Milestones: First Blind Takeoff, Flight and Landing, 1929. Engineering and Technology History Wiki. https://ethw.org/Milestones:First_Blind_Takeoff,_Flight_and_Landing,_1929 [VERIFIED]
  9. ASME International (2000). The Link Flight Trainer: A Historic Mechanical Engineering Landmark, 10 June 2000. https://www.asme.org/wwwasmeorg/media/resourcefiles/aboutasme/who%20we%20are/engineering%20history/landmarks/210-link-c-3-flight-trainer.pdf [VERIFIED]
  10. Binghamton University Libraries (n.d.). Flight Simulators & Other Inventions, in The Life of Edwin A. Link. https://digitalprojects.binghamton.edu/s/edwin-link-exhibit/page/flight-simulators-other-inventions [VERIFIED]
  11. European Association for Aviation Psychology (n.d.). Selection in Aviation: An EAAP Report (v2). https://www.eaap.net/get/10022/selection-in-aviation-v2-pdf.pdf [VERIFIED — the AAF figures it reports are [SECONDARY], drawn from DuBois 1947, Flanagan 1948, Melton 1947, Guilford & Lacey 1947]
  12. Shorrock, S. (2018). Human Factors and Ergonomics: Looking Back to Look Forward. Humanistic Systems, 25 February 2018. https://humanisticsystems.com/2018/02/25/human-factors-and-ergonomics-looking-back-to-look-forward/ [SECONDARY — source of the quoted Fitts & Jones lines]
  13. Historic Croydon Airport (2018). The Gosport System: Robert Smith-Barry and Pilot Training. https://www.historiccroydonairport.org.uk/the-gosport-system-robert-smith-barry-and-pilot-training/ [SECONDARY]
  14. Pew, R. W. (2010). Alphonse Chapanis: Pioneer in the application of psychology to engineering design. APS Observer, 1 April 2010. https://www.psychologicalscience.org/observer/alphonse-chapanis-pioneer-in-the-application-of-psychology-to-engineering-design [SECONDARY]
  15. National Air and Space Museum (2026). Scandalous: The Air Mail Crisis of 1934 (AirSpace S11E8, with Dr Bob van der Linden). https://airandspace.si.edu/stories/editorial/airspace-air-mail-scandal [SECONDARY]

Cited but not obtained (recommended acquisitions)

  • Chapanis, A. (1999). The Chapanis Chronicles: 50 Years of Human Factors Research, Education, and Design. Santa Barbara, CA: Aegean Publishing. ISBN 9780963617897.
  • Smith-Barry, R. (1917). General Methods of Teaching Scout Pilots. Royal Flying Corps.
  • DuBois, P. H. (Ed.) (1947). The Classification Program. AAF Aviation Psychology Program Research Reports No. 2. Washington, DC: US GPO.
  • President’s Task Force on Aircraft Crew Complement (1981). Report. Washington, DC: US Department of Transportation. (Chair: J. L. McLucas.)
  • Daniel Guggenheim Fund for the Promotion of Aeronautics (1930). Equipment Used in Experiments to Solve the Problem of Fog Flying: A Record of the Instruments and Experience of the Fund’s Full Flight Laboratory. https://catalog.hathitrust.org/Record/001039798