Human Factors and Information Processing in Aerospace

Overview of Human Factors and Human Error in Aerospace

  • Human Factors Objective: The primary goal of studying human factors in aviation and aerospace is to mitigate and control the propensity for human operators to act unpredictably or unreliably compared to automated systems, while accounting for the reality that machines also experience failures.

  • Definition of Human Error: Human error is defined as an action taken by a human being that results in an incorrect or incomplete output, meaning the achieved results were neither expected nor desired.

  • Quantitative Breakdown of Factors Contributing to Human Error:

    • Procedural Problems: 23.78%23.78\%

    • Personnel / Human Error: 22.54%22.54\%

    • Equipment / Material Problems: 19.63%19.63\%

    • Training Deficiencies: 15.05%15.05\%

    • Management Oversight: 8.98%8.98\%

    • Design Problems: 8.73%8.73\%

    • External Phenomena: 1.30%1.30\%


Factors Contributing to Human Error
  • Human Fallibility vs. Automation: Modern technologically advanced airframes present high volumes of continuous data, leading to information and input overload. Human pilots must continuously receive, process, and act upon this stream of data throughout all flight phases.

  • Real-World Aviation Case Studies in Human Error and Adaptability:

    • Air France Flight 447 (June 1, 2009):

    • Aircraft & Context: Airbus A330 flying over a stormy region of the Atlantic Ocean.

    • Trigger Event: Sudden loss of airspeed information caused by ice crystals freezing the external pitot tube measurement device.

    • System Response: Although the airframe remained fully functional, the loss of reliable airspeed data caused the autopilot to disengage automatically and relinquish manual control to the flight crew.

    • Human Breakdown: The pilots misinterpreted cockpit alarms and flight displays, failed to maintain effective Crew Resource Management (CRM) and communication, and held pitch-up inputs that induced an aerodynamic stall, resulting in a crash into the Atlantic Ocean and the loss of all 228228 lives on board.

    • Boeing 737 MAX Crashes:

    • Events: Lion Air Flight 610 (October 2018; 189189 fatalities) originating in Indonesia/Bali, and Ethiopian Airlines Flight 302 (March 2019; 157157 fatalities) originating in Ethiopia.

    • System Failure: Improper activation of the automated Maneuvering Characteristics Augmentation System (MCAS) designed to prevent aerodynamic stalls.

    • Human Factor: In both instances, the airframes remained structurally intact, but the crews failed to execute emergency procedures using cutout switches to disable autopilot and automated trim controls, preventing manual recovery.

    • US Airways Flight 1549 ("Miracle on the Hudson" - January 15, 2009):

    • Crew & Context: Captain Chesley "Sully" Sullenberger and Co-pilot Jeffrey Skiles taking off from New York's LaGuardia Airport.

    • Trigger Event: Multiple bird strikes minutes after takeoff resulted in complete dual-engine loss of thrust.

    • Cognitive Challenge: No Standard Operating Procedure (SOP) existed for dual-engine failure at low altitude during initial climb out.

    • Human Outcome: Relying on high-level expert decision-making and cognitive flexibility, the crew successfully ditched the aircraft in the Hudson River, leading to the survival of all 155155 passengers and crew members.

Human Information Processing Models and Theories

  • Core Concept: Human information processing models describe how information enters the cognitive system via sensory organs, undergoes perceptual processing, interacts with memory structures, and culminates in response selection and execution.

  • Four Fundamental Components:

    1. Input: Environmental stimuli gathered via sensory channels.

    2. Processing: Cognitive transformations, data interpretations, and decision-making.

    3. Storage: Interplay between short-term working memory and long-term memory stores.

    4. Output: Motor, physical, or verbal execution of a selected response.

  • Model Properties and Limitations:

    • Models are theoretical, qualitative frameworks that cannot be taken literally as physical representations of brain architecture.

    • Cognitive processing stages do not operate in a strict linear sequence, nor do they possess rigid start or stop points; processing involves continuous feedback loops and iterations.


Wickens' Model of Human Information Processing
  • Wickens' Model of Human Information Processing:

    • A qualitative framework describing stages: sensory store, perception, decision and response selection, response execution, supported by working memory, long-term memory, and attentional resource pools.

    • Each processing stage requires a discrete duration of time, which increases under conditions of uncertainty, high noise, or conflicting cues.

  • Hick-Hyman Law (1952–1953):

    • Establishes a linear relationship between reaction time (RTRT) to a signal and the amount of information present, defined by the number of choice alternatives available.

    • Principle: Reaction time slows logarithmically as the number of available options increases.

    • Cockpit Application: Warning systems should be designed to represent fewer distinct diagnostic categories (e.g., 11 of 44 situation alerts rather than 11 of 88). Empirical research confirms that lower choice density speeds decision-making.

  • Fitts's Law (1954):

    • Describes a linear relationship between movement time (TT) and the target distance (DD) relative to target width (WW).

    • Mathematical Formula:

T=a+b×log2 ⁣ ⁣  ⁣ ⁣2DWT = a + b \times \text{log}_2\negthinspace\negthinspace\text{ }\negthinspace\negthinspace\frac{2D}{W}

  • Principle: Movement time increases as the target distance increases or as the target size decreases.


Fitts's Law Equation and Diagram
  • Design Applications: Vital flight controls or emergency shutoff interfaces are made larger and positioned closer to primary control paths (e.g., heavy machinery emergency stop buttons are large, red, and raised, whereas startup buttons are smaller and recessed green switches).

    • Limitations of Information Processing Frameworks:

  • Quantifying the exact bit-rate of information in dynamic real-world environments is difficult.

  • Human performance depends on extra-information factors including 3D vs. 2D visual perception differences (e.g., evaluating physical visual cues visually vs. interpreting a 2D cockpit display).

  • Choice selection is influenced by risk assessment, potential outcome severe consequences, operator experience, and time pressure constraints.

Aviation Evolution, Core Cognitive Tasks, and Flight Dynamics

  • Historical Shift in Pilot Mental Demands:

    • 1930s Early Aviation: Open-cockpit, single-engine biplanes with direct mechanical flight control. Pilots worked in isolated airspace without air traffic control, two-way radios, or complex traffic density. Tasks were limited to manual flight control, visual navigation, and terrain avoidance.

    • Modern Aerospace Environment: Highly automated, multi-crew platforms operating in dense, controlled airspace. Operators must manage automated flight systems, coordinate with co-pilots, communicate with Air Traffic Control (ATC), and manage ground maintenance directives.

  • The Primary Human Factors Challenge: Optimizing how human attentional resources are allocated, mobilized, and sustained under varying levels of mental workload and environmental dynamism.

  • Three Critical Cognitive Elements Determining Performance:

    1. Perception: Input mechanism for information processing. Susceptible to visual illusions, spatial disorientation, and visual inattention.

    2. Manual Control: Consists of aviating (averting aerodynamic stalls and maintaining control - the highest priority) and navigating (route tracking, weather avoidance, and obstacle clearance - the second priority).

    3. Communication: Managing information exchange between cockpit crew members and ATC to prevent fatal misunderstandings.

  • Six Critical Dimensions of Aircraft Control (Three Spatial Axes):


Aircraft Axes and Dimensions of Movement
  • Lateral Axis (Pitch): Controls nose-up and nose-down pitch attitudes across the lateral plane.

  • Vertical Axis (Yaw): Controls directional heading and yaw rate of change.

  • Longitudinal Axis (Roll): Controls aircraft roll angle along the path, maintaining wing levelness and managing airspeed/groundspeed vectors.

  • Autopilot Interplay: While automated flight systems manage these axes during cruise, pilots must maintain continuous readiness to resume manual control loops during system disengagement or failure.

    • Cockpit Communication Optimization Strategies:

  • Aviation Phonetic Alphabet: Standardization of letter soundings (e.g., Alpha, Bravo, Charlie) and explicit phraseology (e.g., utilizing "negative" instead of "no").

  • Redundancy & Readback Protocols: Structural repetition of flight clearances and command strings between pilots and ATC controllers.

  • Articulation Standards: Enforced unhurried cadence, clear pronunciation, and strict vocabulary limitations.

  • Language Fluency: Mandatory international standard usage of proficient English for commercial aviation ATC operations.

  • Controller-Pilot Data Link (CPDL): An automated digital link that converts spoken ATC voice directives into written cockpit text displays.

    • Performance Finding: Empirical research confirms CPDL increases communication accuracy but does not accelerate comprehension speed.

Performance Drivers: Workload, Speed-Accuracy Trade-Offs, and Situation Awareness

  • Stimulus-Response (S-R) Compatibility: Performance speed and response accuracy increase when physical interface motions match operator mental expectations.

    • Turning a dial clockwise to increase a numerical parameter.

    • Placing a control knob to the right side of a vertical display indicator matching left-to-right reading habits.

  • Hick-Hyman Law Boundary: The Hick-Hyman relationship degrades when cockpit signal and action choices exceed approximately 88 alternatives, rendering classic reaction time equations insufficient for ultra-complex cockpits.

  • Speed-Accuracy Trade-Off: As task performance speed is artificially forced, operator error rate increases systematically.

  • Mental Workload: The dynamic relationship between the finite mental capacity supplied by a human operator and the concurrent mental resource demand imposed by task requirements. Attention serves as the operational energy fueling this processing system.

  • Situation Awareness (SA): Defined as the perception of environmental elements within a defined volume of time and space, the comprehension of their immediate operational meaning, and the projection of their future status in the near term.

  • The Automation/SA Paradox: High cockpit automation significantly reduces physical and routine mental workload for flight crews; however, it removes the pilot from direct manual control loops, leading to reduced Situation Awareness and delayed emergency response times.

Multitasking Dynamics in Aviation Operations

  • Concurrent Multitasking: Executing two or more cognitive or motor tasks simultaneously.

    • Success Factors: Success depends on whether task demands draw upon shared or separate attentional/perceptual modalities.

    • Successfully Multitasked Pairs (Non-competing channels):

    • Tracking a visual glide-slope display (spatial/visual attention) while receiving spoken ATC directions (verbal/auditory attention).

    • Manipulating flight controls and speaking (motor/response output) while scanning exterior airspace for aircraft traffic (perceptual/input processing).

    • Looking forward through the windscreen at external terrain (visual attention) to verify flight path trajectory displayed on cockpit flight instruments (spatial attention).

    • Unsuccessfully Multitasked Pairs (Competing channels):

    • Tuning an instrument frequency panel while concurrently listening to complex verbal ATC clearances.

    • Listening to simultaneous verbal instructions from an ATC controller and a co-pilot (both competing for auditory processing resources).

    • Reading an approach plate/map while visually scanning external sky for visual air traffic (both competing for visual attention).

  • Sequential Multitasking: Interleaving separate tasks by switching focus sequentially from one to another. This strategy is adopted when overall cognitive load exceeds concurrent capacity.

  • Dangers of Sequential Multitasking:

    • Interruption Management: High-priority ongoing flight tasks (aviating, navigating) are unexpectedly interrupted by secondary tasks. Errors occur if the pilot fails to prioritize the ongoing primary flight loop or delays returning to the main task.

    • Task Tunneling (Attentional Fixation): Total cognitive engagement becomes focused on a single isolated task for an extended period, leading to complete neglect of critical primary flight monitoring responsibilities.

  • Aviation Disasters Caused by Task Tunneling and Checklist Interruptions:

    • Eastern Airlines Flight 401 (December 1972):

    • Context: Lockheed L-1011 flying from New York JFK to Miami, FL with 176176 people on board.

    • Sequence: During descent into Miami, the flight crew noticed that the nose landing gear indicator light failed to illuminate. The crew entered a hold over the Florida Everglades at an assigned altitude of 2000 ft2000\,\text{ft} to troubleshoot a suspected light bulb failure.

    • Attentional Breakdown: The entire flight crew fixated on diagnosing the nose gear light bulb. During this process, the autopilot was bumped or disengaged, and the crew suffered total task tunneling. Unaware of the gradual loss of altitude, the aircraft crashed into the Everglades at 227 mph227\,\text{mph}, causing 101101 fatalities.

    • Northwest Airlines Flight 255 (August 1987):

    • Context: McDonnell Douglas MD-80 departing Detroit Metropolitan Airport.

    • Sequence: While performing taxi checklists, the crew was interrupted by ATC with a runway assignment change. Upon resuming the checklist, the crew skipped ahead past the interrupted line item, missing the critical pre-takeoff wing flap and slat configuration check.

    • Outcome: The aircraft took off without flaps deployed, entered an immediate aerodynamic stall, rolled erratically, struck a light pole located 2760 ft2760\,\text{ft} past the runway end, clipped a car rental building roof, and crashed onto an interstate highway. All 66 crew members, 148148 passengers, and 22 motorists on the ground died (156156 total fatalities); one 44-year-old girl survived.

Overcoming Information Processing Limits: Expertise Development

  • Role of Expertise: High domain expertise expands effective working memory processing capacity, optimizes attentional resource allocation, and allows operators to overcome inherent cognitive limitations.

  • Currency Principle: Aviation expertise requires continuous, active deliberate practice. Skills decay without ongoing exposure ("flying an aircraft is not like riding a bicycle").

  • Three Stages of Skill Acquisition (Fitts & Posner Framework):

    1. Cognitive Stage: Declarative knowledge is explicitly utilized to understand task demands, construct operational rules, and set task performance goals.

    2. Associative Stage: Physical and procedural actions are refined through deliberate practice. Performance becomes smoother, execution time decreases, and gross procedural errors are eliminated.

    3. Autonomous Stage: Skills and procedural routines are executed automatically with minimal conscious cognitive effort.

    • Environmental Caveat: Automaticity requires stable, predictable stimulus-response mapping. Sudden anomalies in dynamic aviation settings (e.g., unexpected aerodynamic stall warnings) disrupt automaticity, forcing the pilot back into deliberate cognitive processing.

  • Deliberate Practice & Variability: Simple repetitive practice alone does not yield elite expertise. High-level mastery requires deliberate practice incorporating novel situational variations to build structural cognitive flexibility.

  • Neurological Changes in Experts vs. Novices:


Novice versus Expert Continuum
  • Practice increases neural efficiency, resulting in an overall decrease in total cortical metabolic activation during routine task execution.

  • Experts exhibit selective neural resource deployment: lower activation in primary motor and sensory/visual cortical regions, accompanied by targeted increases in prefrontal cortex activation dedicated to executive control, strategic planning, and attentional coordination.

  • fMRI Evidence (Peres et al., 2000): Comparative neuroimaging study of expert vs. novice French Air Force pilots operating flight simulators across variable speeds:

    • Novices: Displayed widespread, intense neural activation concentrated in the right hemisphere, reflecting high visuospatial processing demands.

    • Experts: Showed suppressed activation in primary visual and motor areas alongside structured activation within the frontal and prefrontal cortex, proving a shift toward higher-order attentional strategy management.

    • Summary Comparison: Experts vs. Novices:

  • Knowledge Architecture: Experts store domain knowledge in highly structured networks accessible via long-term working memory.

  • Pattern Recognition: Experts extract meaningful complex patterns from partial environmental cues rapidly.

  • Situation Awareness: Experts extract diagnostic information faster, maintaining superior situational mental models.

  • Metacognition: Experts possess realistic evaluations of their own knowledge boundaries (knowing what they know vs. what they do not know).

  • Intuitive Decision-Making: According to naturalistic decision-making theories (e.g., Gary Klein), experts routinely rely on intuitive recognition-primed decision mechanisms honed through extensive historical experience.

Risky and Adaptive Decision-Making

  • Risky Decision-Making Definition: Evaluating two or more plausible operational choices where environmental evidence must be evaluated, outcomes cannot be predicted with certainty, and chosen options carry potential for severe or catastrophic harm.

  • Three Sequential Stages of Decision-Making:

    1. Cue Seeking: Selective extraction of environmental cues.

    2. Situation Assessment: Integration of cues with prior domain knowledge to form a diagnostic mental model.

    3. Risky Choice Selection: Selecting an operational action path under uncertainty.

  • Adaptive Decision-Making: The capacity of human operators to dynamically switch decision strategies based on environmental constraints, time pressure severity, and hazard consequences.

  • Systematic Cognitive Decision Biases:

    • Salience Bias: Weighting cues based on their physical brightness, visual prominence, or perceptual intensity rather than their true diagnostic informative value.

    • Anchoring and Confirmation Bias: Anchoring on the initial piece of incoming information (e.g., an early weather report) and subsequently filtering or favoring new cues that confirm the original hypothesis while discounting disconfirming evidence.

    • Framing Bias: When choosing between two negative choices—a guaranteed minor loss versus a risky alternative that offers a chance of avoiding loss alongside a risk of catastrophic loss—operators reframe options positively and favor the risky alternative.

    • Sunk Cost Bias: An extension of framing bias wherein an operator persists with a failing or unsafe course of action due to previous investments of time, effort, financial resources, or emotional commitment.

  • Applied Aviation Scenario: Visual Flight Rules (VFR) Pilot Weather Encounter:

    • Operational Context: A non-instrument-rated VFR pilot (permitted to fly solely by visual external references) encounters worsening cloud cover and reduced flight visibility along the planned route.

    • Bias Manifestation:

    • Salience Bias: The pilot fixates on the immediate visual view looking forward out the windscreen, allowing current visual flight conditions to dominate decision-making while ignoring objective instrument/radar indicators showing severe weather ahead.

    • Anchoring & Confirmation Bias: The pilot anchors on the favorable pre-flight weather brief received prior to departure, assuming clear skies ahead while failing to request updated en-route flight watch briefings.

    • Framing Bias: The pilot frames turning back to the departure airport as a guaranteed failure or loss (delaying the mission and wasting time), leading them to select the risky choice of continuing into degrading weather.

    • Sunk Cost Bias: Having flown most of the route, the pilot continues pushing into severe Instrument Meteorological Conditions (IMC) due to the accumulated time and fuel already expended.

Key Academic References

  • Keebler, J. R., Lazzara, E. H., Wilson, K. A., & Blickensderfer, E. L. (Eds.). (2023). Human Factors in Aviation and Aerospace (3rd ed.). Elsevier, Academic Press.