iq.metronics.one — Non-Clinical Cognitive Assessment

Start

What this is

Multi-domain cognitive assessment: fluid, verbal, quantitative, working memory, processing speed, spatial.
Adaptive testing (CAT) for several subtests with confidence intervals.
Results shown as both IQ-style estimate + CI and percentile + CI.

What this is not

Not WAIS, Stanford-Binet, or any clinical IQ test.
Not a medical/psychological evaluation.
Online tests cannot fully replicate controlled proctoring conditions.

I understand this is non-clinical and I want to continue.

Reproducibility seed (optional)

Use the same seed to reproduce the item order for debugging, CAT checks, or research controls.

Norm packs (optional) - improve calibration

The app is self-contained and works out of the box using a bundled baseline norm model. You can optionally load a custom norm pack (JSON) built from your own sample using the included pipeline.

Research Mode (optional)

Enables advanced exports for research and calibration. No data leaves your browser unless you download it.

Enable Research Mode (shows research exports)

Privacy

Runs and history are stored only in your browser (localStorage). No server. No tracking. You can export your run data at the end.

-

Results

History (this browser)

Quality flags

Tab switches and focus losses: leaving the page during timed items.
Rapid guesses: extremely fast answers that may be random.
Attention checks: simple instructed-response items at the start.

Interpretation notes

Your estimate is reported with uncertainty (95% CI). Online settings add additional noise.
Scores are most useful for self-tracking over time, using the same device and conditions.
Custom norm packs can improve calibration for a specific audience or cohort.