Experiment 2: Stacked Deck

Resonance with Randomness

The Princeton Engineering Anomalies Research (PEAR) lab famously asked whether human intention could bias the outputs of random machines. Their studies collected millions of trials and reported tiny but statistically significant deviations. But follow-up replications by other labs—sometimes using PEAR’s own equipment and protocols—found no such effect.

So why revisit a test that already failed? Because repeating PEAR’s design faithfully was never enough. If the paradigm itself had blind spots, then cloning it would naturally yield nothing. Experiment 2 doesn’t rerun PEAR — it redesigns PEAR.

Replications after PEAR produced null results. But they carried forward the same vulnerabilities:

Experimenter presence: operators and researchers remained in the loop, leaving subtle cues and expectancy effects unguarded.
Timing insensitivity: no one asked whether awareness depends on when decoherence is sampled.
No internal controls: trials were compared only to chance distributions, with no built-in “ghost” measure.
Continuous peeking: results were inspected as they accumulated, opening the door to premature conclusions.
Device assumptions: RNG validation was limited to calibration checks; they didn’t embed continuous health monitoring as part of the live experiment. Ours runs ongoing tests in the background.

Experiment 2 is built to close those gaps using and reframe the question itself. We’re not asking whether awareness forces a machine to behave differently. We’re asking whether awareness can resonate with decohering outcomes, gently biasing which lawful possibilities stabilize depending on where attention rests.

Participate In This Study

Guardrails Against Bias and Error

We’ve made several deliberate departures from the PEAR model and its replications:

Independent entropy sources: one block uses a classical physical RNG (via random.org), the other a quantum RNG.
Paired outcomes: Each RNG call returns a pair of bytes: one primary, one demon. Like Maxwell’s demon in physics, the second byte doesn’t act directly, but it gives us a reference point — a built-in control to compare against the primary.
Blind thresholds: no one looks at results until at least 300 runs are complete, and no conclusions are drawn until 1000+. This prevents early peeking or p-hacking.
No experimenters present: everything is automated. Participants interact with the system directly, removing subtle social cues or expectancy effects.
RNG validation: the random streams are continuously checked against statistical health tests, ensuring the baseline entropy is sound.

Unlike Experiment 1, this study does not use a commit–reveal vault. That’s intentional. Resonance has to be tested on live decoherence events, not on pre-sealed sequences. The safeguards here are designed around live controls and delayed analysis, rather than cryptographic locks.

Two Modes of Testing

We run the experiment in two blocks:

Baseline (PRNG): outcomes from a physical random source, echoing the setup used at Princeton.
Main (QRNG): outcomes from a quantum random number generator. This gives us a benchmark rooted directly in quantum decoherence, letting us ask whether awareness can align with outcomes at the quantum level as well as the classical.

What We’re Looking For

The baseline expectation is 50% success, pure chance. We’re watching for:

Reliable deviations above or below that mark.
Asymmetry between primary and ghost outcomes.
Patterns of alignment across both classical and quantum blocks.

If awareness really resonates with decoherence, we’d expect small but measurable shifts that respect physical law yet show lawful biasing over time.

Interpreting Marginal Results

We learned from PEAR: small effects can vanish under scrutiny. That’s why we’ve designed this experiment to minimize loose ends.

If nothing appears, we can say with confidence it isn’t because of biased experimenters, faulty RNGs, or statistical corner-cutting.
If something does appear, the paired controls and blind thresholds give us reason to take it seriously.

Marginal signals matter only if they repeat under better conditions. This design is about building those conditions.

Closing Thought

PEAR asked whether minds could sway machines. Their replications suggested the answer was no. We’re asking a subtler question: can awareness gently resonate with decoherence, shaping probability not by force but by alignment?

By rebuilding the experiment with demon under the hood outcomes, blind thresholds, quantum sources, and no experimenters, we aim to give this old question a new and sturdier frame.

Methods Appendix

Trial structure: single-button presses across two blocks.
Block A (PRNG): outcomes from a physical random source (random.org).
Block B (QRNG): outcomes from a true quantum RNG.
Every RNG call returns a pair of outcomes. One is designated primary, the other demon. Their roles alternate each trial, so participants can’t know which is in play.
Scoring: binomial test against p = 0.5 baseline.
Demon controls: demon stream analyzed in parallel for comparison.
Blind thresholds: results remain unseen until ≥300 runs; no analysis finalized until ≥1000 runs.
No experimenters present: sessions are fully participant-driven.
RNG validation: statistical health checks run continuously.
Priming: an additional variable is embedded but not disclosed until after results, to avoid expectation bias.

Next Phase: Timing and Emotion

The app scaffolds several timing arms to probe whether resonance depends on when decoherence is sampled:

Open (default): immediate sampling.
Scramble: inserts random micro-delays, destroying phase alignment.
Synced: snaps sampling to a steady rhythm (e.g. 60 Hz), potentially amplifying resonance.
Blind (future): combines commit–reveal sealing with resonance testing.

At present, all runs use the open arm. The others are reserved for future tests.

Another extension is emotional coherence: adding biometric measures (HRV, EEG, etc.) to test whether emotionally stable intention correlates with outcome shifts. Together, timing and emotion will sharpen the resonance hypothesis into something more directly testable.