η̂ₜ
Estimated adaptive evidence-integration rate.
Published September 6, 2026 · v0.2
Homo sapiens
How can we test whether an adaptive system can still use valid corrective evidence to update its internal model and adjust its response?
01 / Published Research
The research question — without diagnostic shorthand
BenchEWS Individual investigates the conditions under which an observing adaptive system loses the capacity to use valid corrective evidence to update its internal model and adapt its response — and whether changes in this process can be measured over time.
Status boundary: BenchEWS Individual is a published theoretical research line within the BenchEWS Research Programme. It provides a falsifiable hypothesis and methods framework for studying time-resolved changes in adaptive evidence integration, prediction-error persistence, and candidate behavioral lock-in dynamics.
02 / Mathematics
An exact decomposition, not a universal shortcut
The earlier shorthand AC1(PE) ≈ 1 − η is not a general identity. The published v0.2 uses the exact decomposition:
PEₜ₊₁ = (Oₜ₊₁ − Oₜ) + (1 − ηₜ)PEₜPEₜ₊₁ = (1 − ηₜ)PEₜ + (εₜ₊₁ − εₜ)Under a locally near-stationary environment, the second form follows. The composite noise term is generally not independent of the current prediction error. Persistence must therefore be tested as model-, regime-, and condition-dependent. ρ₁(PE) = −η/2 applies only to the fully specified minimal model.
Open the mathematical boundary →03 / Measurement Architecture
Minimal measurement vector
Four measurement slots structure a possible time-resolved test. The vector is published but has not been empirically validated.
BI(t) = [ η̂ₜ, Persistence(PE)ₜ, I(S;R)ₜ, Trecovery ]Persistence(PE)ₜ
Abstract slot for prediction-error persistence; AC1 is a primary candidate, not a universally fixed metric.
I(S;R)ₜ
Rolling environment–response mutual information; a normalized robustness measure such as NMI may supplement it.
T_recovery
Time or number of trials required to return to a defined baseline after perturbation.
04 / Falsifiability & Estimator Latency
Ten hard falsification criteria
The framework specifies ten explicit conditions under which its central temporal-lead hypothesis should be considered unsupported or falsified. These include null and competing models, parameter recovery, robustness to window selection, and correction for estimator-specific delays.
Observed lead is not necessarily genuine lead
Measures use different window lengths and detection delays. An apparently earlier warning may therefore be created or shifted by the estimator itself. Observed lead and estimator-latency-corrected lead must be reported separately.
05 / Studio 3.0 Research Horizon
From the published framework to planned Studio 3.0 infrastructure
- 01Published: theoretical framework, mathematical derivations, measurement candidates, falsification criteria, and proposed experimental architecture.
- 02Planned: methodological operationalization, experimental design, parameter recovery, estimator-latency tests, and competing-model analysis.
- 03Still open: reproducible Studio 3.0 implementation, real data collection, independent replication, and empirical validation.
Binding separation: BenchEWS Individual v0.2 = published theoretical framework. BenchEWS Studio 3.0 = PLANNED · NOT IMPLEMENTED · NOT VERIFIED.
06 / Research–Tool Boundary
Published framework and reflection tool are not the same measurement model
The published BenchEWS Individual v0.2 paper (DOI 10.5281/zenodo.22477244) defines the theoretical measurement vector B_I(t) shown above. The interactive self-reflection tool below is a separate, deliberately bounded HCI/software artifact using six subjective self-report items. Its ratings, medians and [Q1–Q3] summaries do not operationalize, estimate, validate or substitute for η̂ₜ, Persistence(PE)ₜ, I(S;R)ₜ or T_recovery.
Evidence status: Paper v0.2 = published theoretical and empirically testable framework, with no empirical data reported. Reflection tool = experimental descriptive self-observation, not an empirical validation of the paper and not a personal early-warning system.
Open the authoritative published paper ↗EXPERIMENTAL · SELF-REFLECTION · NON-DIAGNOSTIC
BenchEWS Individual · N-of-1 self-observation
The six dimensions are shown separately as subjective self-reports. There is no global score, personal EWS forecast, diagnosis, or risk classification.
Personal time-series values entered here are processed only in this module's local browser state. This statement concerns this tool's input data, not unrelated website infrastructure.
1 · Enter2 · Describe locally3 · View raw trajectory4 · Reflect qualitatively
GUIDE / FAQ
Four questions: How to use this tool
Open four questions and answers
1 · What do I enter?
The actual observation date and time, plus six subjective ratings from 0 to 10. Add observations at later times.
2 · What do I get?
Dated raw values and trajectories for each dimension; from five observations, median, middle 50%, minimum/maximum and MAD. No personal forecast.
3 · What is the benefit?
Notice changes in your own reports and formulate testable questions. Temporal co-occurrence does not establish causality.
4 · How is this linked to BenchEWS?
BenchEWS researches dynamical early-warning indicators. This tool first provides an exploratory N-of-1 observation basis; AR(1), recovery and personal EWS are neither computed nor validated.
00 / Measurement model & provenance
Measurement model & provenance
The six inputs are single, self-defined self-report items. They are not validated psychometric scales, latent constructs, or objective sensor measurements.
- Scale type: subjective 0–10 ratings with semantic anchors; interval scaling is explicitly not claimed.
- Every dimension remains separate; no factor, trait, risk, or diagnostic inference is made.
- Provenance: SELF-REPORT → LOCAL CALCULATION → DESCRIPTIVE → QUALITATIVE REFLECTION.
- Reliability, factor structure, construct validity, and criterion validity are not established for these six items.
- The display therefore uses median and middle-50% range [Q1–Q3], not Pearson lag-1, OLS trend, or global aggregation.
SELF-REPORTLOCAL CALCULATIONDESCRIPTIVENON-DIAGNOSTIC
Current input – not saved yet. Only “Add observation” includes these values in the analysis.
Timestamps are retained for provenance. The public tool currently deliberately omits interval-dependent metrics such as Pearson lag-1, recovery time, and OLS trend.
01 / Local descriptive calculation
Descriptive metrics by dimension
Every dimension remains separate. From five observations onward, robust descriptive metrics are shown: median, [Q1–Q3], minimum/maximum, and MAD. There is no weighting, inversion, ranking, or person score.
N0observations
Software abstention
- N < 5: show raw data only.
- N ≥ 5: show median, [Q1–Q3], minimum/maximum, and MAD for each dimension.
- This boundary is a conservative UI rule, not a validated scientific minimum N.
- No silent interpolation and no lag-1, EWS, CSD, recovery, coupling, or peak-learning inference.
N-of-1 research workbench
Descriptive analysis of actually observed timestamps. No interpolation, no global score, and no interval-dependent EWS inference.
- Observation basis
- N = 0
- Observation period
- —
- Typical sampling interval
- —
- Largest gap
- —
Perceived tension
Latest saved value: —- Median
- not calculated
- Middle 50% [Q1–Q3]
- not calculated
- Minimum / Maximum
- not calculated
- MAD
- not calculated
Self-reported available energy
Latest saved value: —- Median
- not calculated
- Middle 50% [Q1–Q3]
- not calculated
- Minimum / Maximum
- not calculated
- MAD
- not calculated
Perceived decision pressure
Latest saved value: —- Median
- not calculated
- Middle 50% [Q1–Q3]
- not calculated
- Minimum / Maximum
- not calculated
- MAD
- not calculated
Perceived recovery
Latest saved value: —- Median
- not calculated
- Middle 50% [Q1–Q3]
- not calculated
- Minimum / Maximum
- not calculated
- MAD
- not calculated
Self-reported parallel obligations
Latest saved value: —- Median
- not calculated
- Middle 50% [Q1–Q3]
- not calculated
- Minimum / Maximum
- not calculated
- MAD
- not calculated
Self-reported concentration
Latest saved value: —- Median
- not calculated
- Middle 50% [Q1–Q3]
- not calculated
- Minimum / Maximum
- not calculated
- MAD
- not calculated
02 / Short description
SELF-REPORTDESCRIPTIVEN = 0NO RANKING
Below five observations, only the raw trajectory is shown.
Normal variation can look like change. Frequent checking can influence self-perception; this tool is not designed for alerting or continuous self-surveillance.
This tool does not measure resilience, burnout, mental illness, or collapse probability. It provides no medical, psychiatric, employment, insurance, educational, legal, or moral assessment and no automated recommendation.Raw data
No observations yet.
03 / Evidence Lab
Evidence lab
Mathematical self-observation and symbolic interpretation systems are structurally separated. They are not evidence for each other and are never combined.
Measurement / calculation layer
BenchEWS · Self-observation
SELF-REPORT / CALCULATED
Raw values, median, [Q1–Q3], min/max and MAD
not validated as individual diagnostics
Epistemic firewall: symbolic systems do not influence BenchEWS self-observation inputs, calculations, metrics, or generated text.
Symbolic / interpretive comparison layer
Archetypal interpretive framework
no validated individual timeline forecastBirth data and traditional interpretive rules
personal predictive effects are not empirically validated hereEsoteric synthesis framework
not validated as a predictive personal modelCalendar and birth constellations
individual effects are not empirically validated hereMathematically rendered symbolic time curve
calculable curve ≠ measured system state04 / Qualitative reflection
- Which external events could explain an observed trajectory?
- Did the personal meaning of the scale change during observation?
- Are entries missing specifically on highly demanding days?
- Which statement remains if every symbolic interpretation is removed?
This tool does not measure resilience, burnout, mental illness, or collapse probability. It provides no medical, psychiatric, employment, insurance, educational, legal, or moral assessment and no automated recommendation.
Not for high-stakes decisions
This tool must not be used for hiring, employee monitoring, performance evaluation, insurance, credit or lending decisions, school or university placement, immigration or asylum decisions, child-custody or family-law decisions, security-clearance decisions, law enforcement, government profiling, use in legal proceedings, clinical diagnosis, or any other decision affecting a person's rights, access, employment, or care.
07 / Ethical Boundaries
Binding ethical boundary
BenchEWS Individual studies time-resolved processes within defined observational and experimental regimes. The framework does not support moral, diagnostic, or global judgments about a person.
- no psychiatric or medical diagnosis
- no automated personality, ideology, or employee assessment
- no credit, insurance, law-enforcement, or government profiling decisions
- no covert surveillance or moral ranking
08 / Cross-scale Research
Two scales of the same overarching question
Quality-Driven Propagation & Adaptive Reopening addresses networked information systems; BenchEWS Individual addresses individual adaptive systems. Both ask when adaptive systems lose correction pathways and how that loss might become measurable. They do not constitute an empirically confirmed unified theory.
08 / Related published research