Does training show up blind?.

As a taster’s vocabulary grows, do their blind results change?

Status: Open
Protocol: Version 1.0, 2026-10-18
Consent: S.training.v1 (earlier data: S.training.v1.past)
Investigator: John Schuster, Drunken Unicorn Productions
Preregistration: Being posted to the Open Science Framework; the link will appear here
Ethics review: Not yet reviewed; we intend independent review before any results are published
Target: 34 tasters with at least 54 rounds
Runs until: The target is reached, or 2029-10-18

Why this study

Smell-training research has mostly studied people recovering lost smell, and the evidence that training changes a healthy adult’s detection threshold is weak. Training plausibly changes something else: what a person can name and tell apart. Savorant records both sides, unprompted vocabulary and blind discrimination, for the same people over time, which no published study has done for whiskey.

What taking part means

Nothing extra. Keep tasting and running triangle rounds the way you already do. Your triangle rounds, and the words you name on your own in tasting sessions (how many distinct ones, and when).

To join: in the app, turn on research contribution in Settings → Research, then open Does training show up blind? under Research studies, read it, and agree. You can leave with one tap.

Hypotheses

H1 (primary). Triangle accuracy is higher in rounds 37–54 than in rounds 1–18.
H2 (secondary). Growth in unaided vocabulary predicts triangle accuracy beyond the number of rounds already completed.

Design

Observational, longitudinal, within-taster. No control group.

Who can take part. Adults 21 or older in the United States who use Savorant, have research contribution turned on, and have joined this study in the app.

Outcomes

Primary. Each taster’s change in proportion correct between their first 18 triangle rounds and rounds 37–54.

Secondary and exploratory.

Mixed-effects logistic regression: correct ~ log(distinct unaided words) + rounds completed, random intercepts for taster and pair, random slope for taster.
The same analysis restricted to rounds on a single repeated pair (sensitivity).

Sample size

34 tasters with at least 54 rounds (paired t-test, d = 0.5, α = 0.05 two-sided, power 0.81: a 10-point change against an assumed standard deviation of change of 0.20, the binomial spread of two 18-round blocks plus between-taster variation). Recruitment target 50, allowing for 30% who stop early.

Analysis

Primary: two-sided paired t-test on per-taster change (α = 0.05), with a Wilcoxon signed-rank test as sensitivity analysis. Effect reported as mean change with 95% CI.

Data filed before a taster joined is used only if they also agree to the separate earlier-data question. It is labeled as earlier data. The primary analysis uses only data filed after joining; the same analysis including earlier data is reported as a sensitivity analysis.

When a study reports more than one secondary test, p-values are adjusted with the Holm method.

Limitations we know about

Without a control group, practice with the test itself, choosing easier pairs over time, and regression to the mean cannot be fully separated from training. The pair random effect and the single-pair sensitivity analysis reduce, but do not remove, these. Any result is association, not proof of cause.

Standards every Savorant study follows

Preregistration. This page is the protocol of record, versioned and dated, and is posted to the Open Science Framework (OSF) before any study data is analyzed. Hypotheses, outcomes, sample size and analysis are fixed here in advance.
Primary and secondary outcomes. Each study names one primary outcome, tested at the stated alpha. Everything else is labeled secondary or exploratory and reported as such, with effect sizes and 95% confidence intervals rather than bare p-values.
No peeking. No interim analysis of the outcomes. The only checks before the sample size is reached are for data quality (duplicates, impossible values), and they never look at results.
Repeated measures. One taster contributes many rounds, so analyses use mixed-effects models with the taster as a random effect, or per-taster summaries, never pooled rounds treated as independent.
Exact methods for small counts. Triangle results are analyzed against the exact binomial with chance at one in three, the way ISO 4120 sets out the triangle test, adapted here for repeated measures.
Exclusions. Rounds are excluded only for the reasons listed in advance: test accounts, the founder’s accounts, rounds filed in a preview build, and duplicates. Every exclusion is counted and reported.
Missing data. Tasters who stop early are kept for every analysis their data supports. The number who stop, and when, is reported.
Deviations. Any change after data collection starts is logged on this page as a dated amendment with its reason, and reported in the paper.
Reporting. Results are reported following the STROBE guideline for observational studies, with the analysis code published alongside.
Conflict of interest. The study is run by the company that sells the app. Preregistration, publishing regardless of outcome, and public analysis code are how that interest is kept out of the result.

The consent you agree to

This is the exact text shown in the app before you join.

Amendments

None. This is the first version of the protocol.

All open studies · The evidence review · Privacy policy · palate@savorant.com