Violema

For physicians and researchers

Help us find
where the reasoning breaks.

A first look at Violema through PsychDoc, our first scientific application. Start with one claim, follow the evidence, and challenge the interpretation.

Allow about 15 minutes. The goal is to identify corrections and useful next questions. Independent source and method review is still open.

Start with the claim

A short review path.

  1. Read the claim and its limit

    The saved result pools mixed trial populations and controls. Read what it does and does not establish before considering its practical meaning.

    Open Claim 1 in the evidence lab

  2. Check the source record

    Follow the dataset, saved analysis, published meta-analysis and author code. Check population, comparator, endpoint and time window against the original sources.

    Download the 12 trial records

  3. Challenge an assumption

    Move the assumed additive shift. The estimate and both interval endpoints move together; no model is refitted and no actual bias is estimated.

    Inspect the canonical analysis

  4. Check what transfers to humans

    Compare the scope of the mechanistic evidence with the clinical claims. Use the animal and human study labels, then inspect the original paper.

    Review the animal / human boundary

PsychDoc specimen

Saved result · independent review pending

Source-linked claim

The saved primary analysis pools 12 trials: Hedges’ g −0.90, with a 95% confidence interval from −1.26 to −0.55.

What the claim does not establish

Negative values favour psilocybin. The dataset combines several clinical populations and controls; this is not an MDD-only effect, a durability estimate or a treatment recommendation.

  1. SYPRES depression-psiloctr, version 25.1.2↗

    Versioned dataset2025sypres-data

  2. PsychDoc Stage B saved sensitivity analysis↗

    Saved analysis · review pending2026psychdoc-stage-b

  3. A living systematic review, meta-analysis and open-data resource of randomized controlled trials of psilocybin treatment for symptoms of depression↗

    Published meta-analysis2026sypres-paper

  4. SYPRES analysis release 26.0.0↗

    Author analysis code2026sypres-code

Saved estimate and interval

  • Saved result
  • Assumed shift

Hedges’ g; Negative values favour psilocybin

Saved analysis with an assumed additive shift−1.50−1.00−0.500.000.50−0.24 reference−0.50 reference

Estimate −0.90; 95% confidence interval −1.26 to −0.55.

The control adds the same assumed shift to the estimate and both interval endpoints. It does not estimate bias, refit the model, or change the interval width.

Reference lines are saved analytical thresholds, not universal clinical cutoffs. A shifted interval crossing a line is conditional on this chosen assumption. Displayed values are rounded; the packet retains full precision.

Prepared source packet PsychDoc 3001296 · SYPRES 25.1.2 · packet 2026-10-10.1Open the record, inspect the limit Inspect

Scope matters

Animal findings.
Human conclusions.

A finding outside one comparison is not absent from the literature.

PsychDoc’s “Where the evidence stops” comparison does not currently include animal findings. Its study-type controls highlight matching evidence in that comparison; they do not search the whole research record.

A separate PsychDoc evidence page includes a mouse finding from Shao et al. (2021). Review that claim in its own experimental context before connecting it to a human mechanism or clinical outcome.

Inspect the comparison Inspect the mouse finding

What would improve this

Five questions
for your review.

A specific correction is more useful than a general endorsement.

  1. Does the claim match its source?

    Flag the exact wording that overstates, omits or misrepresents a finding. Include the publication and passage when possible.

  2. Which differences should block a comparison?

    Identify populations, controls, endpoints, timing or repeated samples that need to be distinguished more clearly.

  3. Is the assumption replay easy to misread?

    Point out where an assumed shift might look like a measured bias, a newly fitted result or a clinical threshold.

  4. Where does the animal-to-human inference go too far?

    Name the missing evidence or experimental condition required to support that step.

  5. What real research decision would this help?

    Suggest one question you would use it for, the minimum evidence needed, and what would make you return to the record.

Bring the correction.
We will keep the trail.

Share short notes with the person who sent this walkthrough: the claim or page, the source, what should change, and why. That feedback can improve the wording, source links, scope labels or next analysis.

Please use published research and general workflow examples. Do not include patient information. This page does not collect review submissions.

The broader method

Another question, different constraints.

The Materials research example keeps source records and test conditions visible without presenting a pooled effect.

Explore the second exampleRead the Violema method