INRAT-H flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

Slip vs. mistake

A slip is an execution failure on a routine, correctly intended action (attention failure). A mistake is a planning failure where the intended action itself is wrong, either rule-based or knowledge-based.

Latent condition

A dormant weakness in design, procedures, training, or maintenance that does not immediately cause harm but can combine with active failures and local triggers to breach defences later.

Performance shaping factor

Any condition that influences human reliability, such as time pressure, workload, fatigue, training, procedures, interface quality, supervision, and environment. PSFs modify error probability estimates.

Violation vs. error

An error is an unintended deviation from intention or a wrong intention. A violation is a deliberate deviation from a rule or procedure, which may be routine, situational, or exceptional.

Working memory limit

Working memory holds a small number of chunks for short periods, commonly cited as about four chunks or a span of seven plus or minus two for simple items.

Multiple resource model

Wickens' model proposes partly separate attentional resources by processing stage, modality, and code. Tasks sharing the same resources interfere more than tasks using different resources.

Signal detection sensitivity

Sensitivity d-prime measures the ability to discriminate signal from noise independent of response bias. It is derived from hit and false-alarm rates.

Vigilance decrement

The decline in detection of rare, low-salience signals over time on watch. It is worsened by monotony, low event rate, and long uninterrupted monitoring periods.

Recognition-primed decision

A naturalistic decision model in which experts match a situation to a recognized pattern and evaluate one option at a time, rather than comparing all alternatives simultaneously.

Loss aversion

A prospect theory finding that losses feel more significant than equivalent gains. It can bias risk decisions, especially when choices are framed as avoiding losses.

Alarm rationalization

A systematic process ensuring each alarm is meaningful, prioritized by consequence and response time, and linked to a defined operator action, reducing flood and nuisance alarms.

Situational awareness levels

Endsley's three levels: perception of elements in the environment, comprehension of their meaning, and projection of future states. Displays should support each level.

Control-display compatibility

The degree to which display and control movements match operator expectations, such as up meaning increase. Violations increase mapping errors and slips.

Redundant coding

Using more than one cue (colour plus shape, position, or text) to convey information. It supports colour-vision-deficient users and reduces confusion when one cue is missed.

Hierarchical task analysis

A method that decomposes a task into goals, subgoals, and operations, showing plans and sequence. It identifies steps where error or workload is concentrated.

Cognitive task analysis

A family of methods that captures the knowledge, decisions, and mental strategies experts use, beyond observable actions. It supports procedure and interface design.

Verification vs. validation

Verification confirms a design meets specified requirements. Validation confirms it supports real task performance under realistic conditions with representative users and scenarios.

NASA-TLX

A subjective workload rating tool with six subscales: mental demand, physical demand, temporal demand, performance, effort, and frustration. It yields a weighted or raw workload score.

Circadian low points

Periods of lowest alertness, typically around 02:00-06:00 and 14:00-16:00. Performance and vigilance are most impaired during these windows, especially on night shift.

Forward shift rotation

A rotation sequence moving day to evening to night, which delays sleep timing and generally aligns better with circadian rhythms than backward rotation.

Sleep debt

Cumulative performance deficit from restricted or disrupted sleep. It builds over days and requires adequate recovery sleep; chronic restriction produces persistent impairment.

Closed-loop communication

A communication protocol where the sender transmits a message, the receiver repeats it back, and the sender confirms. It reduces misunderstanding and verifies shared understanding.

Shared mental model

A common understanding among team members of the task, roles, and situation. It supports coordination and is built through briefings, updates, and cross-monitoring.

Authority gradient

The perceived power difference between team members. A steep gradient can suppress speaking up; mitigation includes assertiveness training and explicit invitation of concerns.

Cross-monitoring

Team members monitoring each other's actions and decisions to catch errors. It provides independent detection but degrades when workload is high or roles are unclear.

Just culture

A safety culture approach distinguishing human error (support), at-risk behaviour (coach), and reckless behaviour (sanction). It encourages reporting by removing blame for honest error.

Swiss cheese model

Reason's accident model in which defences have holes; harm occurs when holes in multiple layers align. It emphasizes latent conditions and organizational factors.

Human reliability analysis

Methods such as THERP, HEART, and CREAM that estimate human error probability for task steps and identify recovery paths. Results are estimates with uncertainty.

Barrier analysis

An investigation method that identifies intended defences against a hazard and assesses why each failed or succeeded, focusing on conditions rather than individual blame.

Training needs analysis

A front-end process identifying task requirements, knowledge and skill gaps, and training objectives before designing content, ensuring training targets real performance needs.

Simulator fidelity

The degree to which a simulator matches the real task. Physical fidelity concerns appearance; functional fidelity concerns behaviour. Fidelity should match the training objective.

Recurrent training

Periodic refresher training required to maintain competence and counter skill decay. Infrequently used emergency skills decay fastest and need more frequent practice.