ECSS Human Factors — Cognitive Performance and Fatigue Assessment (space-systems/ecss/e1011-cognitive-perf)
Use when the task is to apply cognitive performance and fatigue reference
data from ECSS-E-ST-10-11 §4.5.4 to determine whether a crew member is
within safe cognitive limits for a given task -- evaluating sustained
attention duration, working memory loading, and decision-making time
against fatigue-adjusted thresholds, and producing a per-domain and
aggregate risk rating.
Domain quick reference
- §4.5.4 provides reference data for three cognitive performance domains:
sustained attention (the ability to maintain focus on a task for a
continuous period), working memory (the number of information items that
can be held and processed simultaneously), and decision-making (the time
needed to evaluate and select among a set of options). Each domain has a
baseline limit at zero fatigue; the limit degrades as fatigue level
increases.
- Fatigue is sourced from one of three categories: acute task-induced load
(cognitive effort accumulated during a work period), sleep deprivation
(reduced total sleep time below the crew rest requirement), or circadian
disruption (shift of the sleep-wake cycle by duty scheduling). The
category does not change the per-domain threshold formula, but it does
determine which scheduling or operational countermeasure applies and must
be recorded to support root-cause traceability.
- Risk is rated per domain as acceptable (within adjusted limit), marginal
(up to the marginal tolerance band beyond the adjusted limit), or
unacceptable (beyond the marginal band). Aggregate crew cognitive risk is
the worst rating across all assessed domains. A single unacceptable domain
makes the overall assessment unacceptable regardless of other domain
ratings.
- A missing fatigue level or an option count of zero is an incomplete input
and must be rejected before the assessment proceeds; the tool does not
default these to zero or best-case values.
Workflow
- Identify the fatigue level for the crew member being assessed (a value
on the 0.0 to 1.0 scale, where 0.0 is fully rested and 1.0 is the
maximum operationally permitted fatigue level). Reject any value outside
this range before proceeding.
- Categorize each fatigue source contributing to the crew member's fatigue
as acute_task, sleep_deprivation, or circadian. Reject an unrecognized
source type; record the category for scheduling countermeasure tracing.
- For each cognitive domain being assessed, compute the fatigue-adjusted
performance limit:
- Attention: available duration decreases by a fixed amount per 0.1
fatigue unit, with a floor at the minimum useful duration (below
which the task is always unacceptable).
- Working memory: available item capacity decreases by one item per 0.2
fatigue unit, with a floor at the critical minimum capacity.
- Decision-making: required time per option scales up linearly with
fatigue; the total required time is option count times the
fatigue-adjusted per-option time.
- Rate each domain:
- Acceptable: the task demand is at or below the adjusted limit.
- Marginal: the task demand exceeds the adjusted limit but is within the
marginal tolerance band (up to twice the adjusted limit for attention
and memory; up to 133% of the required decision time).
- Unacceptable: the task demand exceeds the marginal band.
- Aggregate domain ratings to the worst per-crew-member risk level.
Record all findings (domain, adjusted limit, demand, rating) for any
domain rated marginal or unacceptable.
- Report the aggregate risk and findings list; proceed with task assignment
only when the aggregate risk is acceptable.
Pitfalls
- Applying baseline limits without fatigue adjustment and reading "within
limit" as acceptable -- the adjusted limit is the operative threshold; a
rested-crew limit applied to a fatigued crew overstates capacity and
understates risk.
- Treating a missing fatigue level as zero fatigue -- an unknown fatigue
state is not the same as a rested state; it is an incomplete input that
blocks the assessment.
- Collapsing marginal into unacceptable or acceptable -- the marginal band
exists so the task can be redesigned (shorter duration, fewer items,
longer decision window) rather than immediately rejected or silently
accepted; flattening it loses the mitigation signal.
- Ignoring fatigue source category once the level is known -- the level
drives the threshold, but the category determines the countermeasure
(rest break for acute load, sleep opportunity for deprivation, schedule
realignment for circadian disruption); dropping the category creates a
finding with no actionable corrective path.
- Assessing only one domain for a multi-domain task -- a procedure that
simultaneously requires sustained attention, high-item working memory,
and rapid multi-option decisions must be assessed across all three
domains; the aggregate risk covers the worst case.
Behavior contract (gate 3)
The fatigue-source categorization, per-domain limit calculation, per-domain
risk rating, aggregate risk, and crew cognitive assessment logic is
exercised by the gate 3 contract test:
scripts/test_e1011_cognitive_perf.py against
scripts/e1011_cognitive_perf_logic.py (stdlib unittest, offline). Run:
python3 scripts/test_e1011_cognitive_perf.py
Compliance
- ECSS standards are freely downloadable (ESA); cite the source and
paraphrase per standards-map.yaml.
- compliance: STANDARDS-REF, gated: false.
1---2name: e1011-cognitive-perf3description: "Use when apply cognitive performance and fatigue reference data under4 ECSS-E-ST-10-11 §4.5.4 to assess crew task feasibility: categorize each fatigue5 source as task-induced, sleep-deprivation, or circadian; determine the6 fatigue-adjusted sustained-attention duration available; evaluate working memory7 loading against fatigue-reduced capacity; and verify that decision time available8 is sufficient for the option count at the crew member's fatigue level. Compute a9 per-domain risk rating (acceptable, marginal, unacceptable) and derive an aggregate10 crew cognitive risk. Flag any domain that breaches its threshold as a finding before11 task or crew assignment proceeds. Trigger: ecss, e-st-10-system-scope,12 cognitive-performance, fatigue, attention, working-memory, decision-making,13 crew-performance."14license: Apache-2.015compliance: STANDARDS-REF16standards:17 - id: ecss18 reference-only: true19gated: false20domain: space-systems21pack: space-systems22compatibility: "agentskills.io SKILL.md; any SKILL.md host (Claude Code, Hermes, OpenClaw)"23metadata:24 source: https://skillmd.com/skills/ashfordeou/e1011-cognitive-perf25 domain: space-systems26 subdomain: ecss27 tags: [ecss, e-st-10-system-scope, cognitive-performance, fatigue, attention, working-memory, decision-making, crew-performance]28 version: 0.1.029 author: Aero Agent Skills30---3132# ECSS Human Factors — Cognitive Performance and Fatigue Assessment (space-systems/ecss/e1011-cognitive-perf)3334Use when the task is to apply cognitive performance and fatigue reference35data from ECSS-E-ST-10-11 §4.5.4 to determine whether a crew member is36within safe cognitive limits for a given task -- evaluating sustained37attention duration, working memory loading, and decision-making time38against fatigue-adjusted thresholds, and producing a per-domain and39aggregate risk rating.4041## Domain quick reference4243- §4.5.4 provides reference data for three cognitive performance domains:44 sustained attention (the ability to maintain focus on a task for a45 continuous period), working memory (the number of information items that46 can be held and processed simultaneously), and decision-making (the time47 needed to evaluate and select among a set of options). Each domain has a48 baseline limit at zero fatigue; the limit degrades as fatigue level49 increases.50- Fatigue is sourced from one of three categories: acute task-induced load51 (cognitive effort accumulated during a work period), sleep deprivation52 (reduced total sleep time below the crew rest requirement), or circadian53 disruption (shift of the sleep-wake cycle by duty scheduling). The54 category does not change the per-domain threshold formula, but it does55 determine which scheduling or operational countermeasure applies and must56 be recorded to support root-cause traceability.57- Risk is rated per domain as acceptable (within adjusted limit), marginal58 (up to the marginal tolerance band beyond the adjusted limit), or59 unacceptable (beyond the marginal band). Aggregate crew cognitive risk is60 the worst rating across all assessed domains. A single unacceptable domain61 makes the overall assessment unacceptable regardless of other domain62 ratings.63- A missing fatigue level or an option count of zero is an incomplete input64 and must be rejected before the assessment proceeds; the tool does not65 default these to zero or best-case values.6667## Workflow68691. Identify the fatigue level for the crew member being assessed (a value70 on the 0.0 to 1.0 scale, where 0.0 is fully rested and 1.0 is the71 maximum operationally permitted fatigue level). Reject any value outside72 this range before proceeding.732. Categorize each fatigue source contributing to the crew member's fatigue74 as acute_task, sleep_deprivation, or circadian. Reject an unrecognized75 source type; record the category for scheduling countermeasure tracing.763. For each cognitive domain being assessed, compute the fatigue-adjusted77 performance limit:78 - Attention: available duration decreases by a fixed amount per 0.179 fatigue unit, with a floor at the minimum useful duration (below80 which the task is always unacceptable).81 - Working memory: available item capacity decreases by one item per 0.282 fatigue unit, with a floor at the critical minimum capacity.83 - Decision-making: required time per option scales up linearly with84 fatigue; the total required time is option count times the85 fatigue-adjusted per-option time.864. Rate each domain:87 - Acceptable: the task demand is at or below the adjusted limit.88 - Marginal: the task demand exceeds the adjusted limit but is within the89 marginal tolerance band (up to twice the adjusted limit for attention90 and memory; up to 133% of the required decision time).91 - Unacceptable: the task demand exceeds the marginal band.925. Aggregate domain ratings to the worst per-crew-member risk level.93 Record all findings (domain, adjusted limit, demand, rating) for any94 domain rated marginal or unacceptable.956. Report the aggregate risk and findings list; proceed with task assignment96 only when the aggregate risk is acceptable.9798## Pitfalls99100- Applying baseline limits without fatigue adjustment and reading "within101 limit" as acceptable -- the adjusted limit is the operative threshold; a102 rested-crew limit applied to a fatigued crew overstates capacity and103 understates risk.104- Treating a missing fatigue level as zero fatigue -- an unknown fatigue105 state is not the same as a rested state; it is an incomplete input that106 blocks the assessment.107- Collapsing marginal into unacceptable or acceptable -- the marginal band108 exists so the task can be redesigned (shorter duration, fewer items,109 longer decision window) rather than immediately rejected or silently110 accepted; flattening it loses the mitigation signal.111- Ignoring fatigue source category once the level is known -- the level112 drives the threshold, but the category determines the countermeasure113 (rest break for acute load, sleep opportunity for deprivation, schedule114 realignment for circadian disruption); dropping the category creates a115 finding with no actionable corrective path.116- Assessing only one domain for a multi-domain task -- a procedure that117 simultaneously requires sustained attention, high-item working memory,118 and rapid multi-option decisions must be assessed across all three119 domains; the aggregate risk covers the worst case.120121## Behavior contract (gate 3)122123The fatigue-source categorization, per-domain limit calculation, per-domain124risk rating, aggregate risk, and crew cognitive assessment logic is125exercised by the gate 3 contract test:126scripts/test_e1011_cognitive_perf.py against127scripts/e1011_cognitive_perf_logic.py (stdlib unittest, offline). Run:128python3 scripts/test_e1011_cognitive_perf.py129130## Compliance131132- ECSS standards are freely downloadable (ESA); cite the source and133 paraphrase per standards-map.yaml.134- compliance: STANDARDS-REF, gated: false.