Skip to main content
A scan can answer at different depths. The same response can be returned as a precise, cortex-wide brain signal, or as a single sentence. You choose the layer that fits the question, and pay attention to how much detail you actually need. The layers run from deepest (most precise, least readable) to simplest (most readable, least detailed).

The layers

Per-vertex fMRI
deepest
The full predicted brain response: activity across roughly 20,000 points on the cortex, for every moment of the content. This is the ground truth. Use it for research, custom modeling, or when you need the unreduced signal.
Per-second networks
layer 2
The cortex-wide response summarized into the 7 brain networks, one value each, every second. Use it to understand why the content performs the way it does, and which systems are firing when.
Per-second lens scores
layer 3
The 7 lenses tracked second by second. Use it to see how attention, emotion, desire, and the rest move across the timeline of your content.
Moments
layer 4
The notable points: the peaks and drops. The seconds where attention spiked, emotion landed, or interest fell off a cliff. Use it to find the exact moments worth keeping or cutting.
Summary
simplest
The headline read: overall score, grade, and a top-line take on each lens. Use it for a fast verdict or to compare versions at a glance.

When to use each

Need a fast verdict?

Use the summary. Score, grade, and lens headlines are enough to compare A vs B or sanity-check before you spend.

Need to find the fix?

Use moments. Jump straight to the second where attention drops or emotion peaks.

Need the full timeline?

Use per-second lens scores to see how every lens rises and falls across the content.

Need to know why?

Use per-second networks to trace a result back to the brain systems driving it.

Per-vertex fMRI

For research and custom modeling, request the unreduced cortex-wide response, the layer everything else is built from.