research / V1 · forced choice

A research platform for anomalous prospection.

Forced-choice trials under cryptographically tamper-evident conditions, contributing to an openly auditable dataset. No overclaiming. No dismissive skepticism. Just clean tools for an open question.

The integrity chain

three steps · pre → sign → reveal
  1. 01

    Pre-commit your choice

    Select between two options before any outcome exists. Your choice is recorded at the moment you click — nothing is generated until after you commit.

    A
    B
  2. 02

    Server-signed timestamp

    The (choice, timestamp) tuple is HMAC-signed server-side before outcome generation. The signature is verifiable proof you chose first.

    HMAC-SHA256
  3. 03

    Generate · record · audit

    The client generates the random outcome via crypto.getRandomValues. The full trial — choice, signature, outcome — is stored with statistical provenance.

Approaching this skeptically?

We cite both supportive meta-analyses (Bem et al. 2015; Storm et al. 2023) and critical replication failures (Wagenmakers 2011; Maier 2020; Walleczek 2025).

Fathom takes no position on the underlying question. The platform exists to enable more rigorous data collection — not to argue a conclusion. Aggregate dataset is public; researcher API is open under a token.

Storm, Tressoldi, Sherwood & Roe
2023 · k = 141, g = 0.02
Walleczek et al.
2025 · N = 26,483, null at high power
Maier et al.
2018 · n = 12,571, BF₀₁ = 10.07
Mossbridge et al.
2024 · RV Exp 2, mixed