Skip to main content
GET
Poll a scan by its scan_id until status is complete. While in flight the response is just { scan_id, status }. When complete you get the rich shape: a per-second timeline of lens scores (the headline capability), plus the detected moments, the buy_moments (where buy intent peaks under the Buy / Not Buy lens), a written summary, and, depending on the options the scan was created with, per-second benchmark, reasons, raw brain-network values, and an fmri_url. A scan that isn’t your org’s, or doesn’t exist, returns 404. The first time you fetch a completed scan, the model charge settles ($2.50 for Qualia, the default and only available model; Mary is temporarily unavailable) and the result is stored to your org’s intelligence.

Path Parameters

string
required
The scan identifier returned by Create a scan, e.g. qualia_run_mq7aht79_hrac1kka.

In-flight response

string
The scan identifier.
string
queued, processing, complete, error, or cancelled. When error/cancelled, an error string is included.

Completed response

string
The scan identifier.
string
complete.
number
The headline 0–100 brain-engagement score for the whole clip, the single number to lead with. This is a directional average over the clip (not a measured ground truth), and it’s the same value the dashboard and result email report. May be null for a clip the model couldn’t score.
string
The letter grade for overall_score (e.g. B, C, D, E). When the model didn’t emit its own grade, it’s derived from overall_score using the standard bands: ≥75 is B, ≥60 is C, ≥45 is D, and anything below that is E.
number
The clip length in seconds. It equals the number of entries in timeline.
string[]
The lenses scored on this scan, in order.
object[]
The per-second breakout, one entry per second of the clip. This is the core of the result: every lens, scored at every second, so you can see exactly where a piece holds and where it slips.
object[]
The notable second-to-second shifts the model flagged: peaks and drops worth a human’s attention. Present when include_reasons.
string
A short written read of the whole clip: the averages and the moments that mattered. Present when include_reasons.
object[]
The buy moments under the Buy / Not Buy lens — the windows where predicted purchase intent (reward minus hesitation, the same basis as the purchase_intent lens) holds above the clip’s average. Each is where you’d place a CTA or the offer. Present when include_reasons; null when no window clears the bar (or the scan has no per-second network data).
object
The full Manipulation read — present only when the manipulation lens was scored on this scan. It’s the same research-grounded block the Manipulation lens renders in the dashboard, and the identical shape returned inline by POST /v1/analyze under result.manipulation. Every number is directional and network-proxied (Yeo-7 networks standing in for the paper ROIs), not a vertex-level fMRI claim.
string
A signed (1-hour) URL to the rendered fMRI artifact, or null. Present only when the scan was created with include_fmri.