Model trace

Select a recorded scenario, compare it with another run, and inspect the C-produced state.

Interactive human cycles model

Scenario
C output

day 1 / 715:05

selected comparison
Recorded daily traces for Baseline rhythmFour aligned plots show circadian wake drive and sleep pressure, available synthetic fuel, hydration, and attention with overload. Straight segments connect discrete ten-minute C model samples.WAKE / SLEEP0—1AVAILABLE FUELkcalHYDRATIONmLATTENTION / OVERLOAD0—1
Stateawake
Available fuel1,454.9 kcal
Hydration1,519.8 mL
Attention0.433
Chosen actionrecover
Combined HR60.8 bpm
Active caffeine96.9 mg
Digestion phase224°

Pre-rendered baseline ready · loading the complete versioned trace.

Keyboard: focus this panel, then ←/→ steps · Space playback · 1—5 inspection depth

Observed data · 41 days

What the record actually supports

derived · single record · non-clinical

Verdict: the claim that this packet contains only a 2-hour and a 20-day cycle does not survive the audit.

~2 hoursnot periodic
The 2-hour lag retains some memory (r=0.184), but a repeating 2-hour sine is not detected (R²=0.014%, p=0.801).
24.848 hourscandidate
The strongest recurring component explains 8.73% of variance (scan-wide p=0.001) and stays within 24.80–24.91 hours under stability checks. Its biological, behavioral, or device source is unresolved.
~20 daysunsupported
A fixed 20-day sine is not supported (p=0.495), worsens blocked prediction, and has only 1.95 repeats in this record. The window is too short to exclude a slow cycle.
Other cyclesbounded result
No remaining 0.5–36-hour peak passed after candidate removal (p=0.588), and no 3–30-day candidate passed its null. That means “not detected here,” not “absent.”

Input cleaning. Of 6,889 unique fresh HR readings, 1,129 readings from 20–34 bpm — including 171 at exactly 20 bpm — were treated as PPG dropout artifacts and excluded. The audit used 5,760 readings across 40 usable daily means; gaps were not interpolated.

Claim boundary. The null tests retain the real observation slots. This describes one wearable packet; it does not diagnose a person or calibrate the synthetic C model above.

Model scope and build details

Scope

This is a systems sketch, not a digital person. It is useful for inspecting feedback, conservation ledgers, and controlled counterfactuals.

Implemented

  • Deterministic fixed-step C state machine
  • Synthetic kcal and mL conservation ledgers
  • Seeded action competition and body-event queues
  • Five controlled schedule scenarios

Not claimed

  • Biological or clinical validity
  • Diagnosis or treatment response
  • Individual calibration or prediction
  • Cross-platform floating-point identity

Build evidence

Schema
human-cycles-showcase-snapshot-v2
Payload
1371bf56ed…e26a2484
C executable
25408481e9…e0c5d2d9
Compiler
cc (GCC) 16.1.1 20260625
Build flags
-O2 -std=c99 -Wall -Wextra -Werror -pedantic
Interpretation
integrity — not biology