Stateless Orthogonal Autonomic Shear (Ω_cal) for Real-Time Cognitive Overload Detection: Cross-Domain Validation from Non-Equilibrium Information Geometry to Deep-Space Flight-Crew Telemetry

Conventional autonomic workload and stress monitors rely on sliding-window heart-rate variability (HRV) averages, such as the root mean square of successive differences (RMSSD) or static heart-rate thresholds. In operational environments characterised by noisy physiological baselines, rapid escalation and severe penalties for false alarms that ranging from acute sensory overload in neurodivergent individuals to autonomous crew health monitoring under 22-minute deep-space communication latencies; windowed HRV metrics exhibit critical diagnostic lag and fail under baseline non-stationarity. Here, we present Sentinel Settle, a local-first autonomic workload and cognitive-overload early-warning architecture adapted from a scale-invariant kinematic trajectory auditor originally developed for non-equilibrium information geometry. By re-anchoring a composite State Uncertainty Coordinate to each individual's initial 30 beat-to-beat R-R intervals and decomposing instantaneous kinematic acceleration into tangential physical exertion versus transverse orthogonal shear gated by a 5-beat tactical filter, the engine isolates autonomic destabilisation without cloud inference or fixed population baselines. Validated on 1.31 hours of 256 Hz NASA Langley flight-crew electrocardiography (25.zip) under a strict 180-second tactical scoring horizon, achieved 18/19 (94.7%) event recall at the Metastable Thinning boundary and 19/19 (100.0%) recall at the Phase Veto boundary , more than doubling standard rolling RMSSD (8/19, 42.1%) and outperforming heart-rate spikes (HR +10 bpm, 13/19, 68.4%). In the 25_LOFT flight scenario, provided 677.7 seconds (~11.3 minutes) of early warning prior to labelled workload onset, while 4-fold Leave-One-Scenario-Out (LOSO) hold-out testing with LOFT completely withheld retained 5/6 event recall under fixed un-tuned thresholds (677.7 s lead time) and 6/6 recall under out-of-sample Youden-calibrated thresholds (590.1 s lead time). Supporting multi-biomarker validation on NASA OSDR Study OSD-575 (SpaceX Inspiration4 9-plex cardiovascular serum panels across N = 4 crew subjects and 28 timepoints) achieved 1.000 sensitivity and 1.000 specificity at . We outline a prospective validation protocol for human analog campaigns (NASA HERA and 6° Head-Down Tilt Bed Rest) to quantify closed-loop haptic pacing recovery slopes .

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-19
DOI
https://doi.org/10.5281/zenodo.24355891
Primary Topic
Heart Rate Variability and Autonomic Control
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Stateless Orthogonal Autonomic Shear (Ω_cal) for Real-Time Cognitive Overload Detection: Cross-Domain Validation from Non-Equilibrium Information Geometry to Deep-Space Flight-Crew Telemetry

Remi Olaiya
1 links
Zenodo (CERN European Organization for Nuclear Research)
Heart Rate Variability and Autonomic Control
preprint

Stateless Orthogonal Autonomic Shear (Ω_cal) for Real-Time Cognitive Overload Detection: Cross-Domain Validation from Non-Equilibrium Information Geometry to Deep-Space Flight-Crew Telemetry

Remi Olaiya
preprint en

Abstract

Conventional autonomic workload and stress monitors rely on sliding-window heart-rate variability (HRV) averages, such as the root mean square of successive differences (RMSSD) or static heart-rate thresholds. In operational environments characterised by noisy physiological baselines, rapid escalation and severe penalties for false alarms that ranging from acute sensory overload in neurodivergent individuals to autonomous crew health monitoring under 22-minute deep-space communication latencies; windowed HRV metrics exhibit critical diagnostic lag and fail under baseline non-stationarity. Here, we present Sentinel Settle, a local-first autonomic workload and cognitive-overload early-warning architecture adapted from a scale-invariant kinematic trajectory auditor originally developed for non-equilibrium information geometry. By re-anchoring a composite State Uncertainty Coordinate to each individual's initial 30 beat-to-beat R-R intervals and decomposing instantaneous kinematic acceleration into tangential physical exertion versus transverse orthogonal shear gated by a 5-beat tactical filter, the engine isolates autonomic destabilisation without cloud inference or fixed population baselines. Validated on 1.31 hours of 256 Hz NASA Langley flight-crew electrocardiography (25.zip) under a strict 180-second tactical scoring horizon, achieved 18/19 (94.7%) event recall at the Metastable Thinning boundary and 19/19 (100.0%) recall at the Phase Veto boundary , more than doubling standard rolling RMSSD (8/19, 42.1%) and outperforming heart-rate spikes (HR +10 bpm, 13/19, 68.4%). In the 25_LOFT flight scenario, provided 677.7 seconds (~11.3 minutes) of early warning prior to labelled workload onset, while 4-fold Leave-One-Scenario-Out (LOSO) hold-out testing with LOFT completely withheld retained 5/6 event recall under fixed un-tuned thresholds (677.7 s lead time) and 6/6 recall under out-of-sample Youden-calibrated thresholds (590.1 s lead time). Supporting multi-biomarker validation on NASA OSDR Study OSD-575 (SpaceX Inspiration4 9-plex cardiovascular serum panels across N = 4 crew subjects and 28 timepoints) achieved 1.000 sensitivity and 1.000 specificity at . We outline a prospective validation protocol for human analog campaigns (NASA HERA and 6° Head-Down Tilt Bed Rest) to quantify closed-loop haptic pacing recovery slopes .

Zenodo (CERN European Organization for Nuclear Research)
Good health and well-being
Heart Rate Variability and Autonomic Control
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon