# Claim Map

> Generated from repository source files. Do not edit this page directly; edit the source file listed below and rerun `vp run docs:generate`.

> Source: `research/06-claim-map.md`

## Claim Map

Status: `wiki evidence map`

This page maps candidate manuscript claims to the local evidence chain. It is a navigation layer; `library/EVIDENCE_MATRIX.md` remains the claim-status source of truth.

For detailed claim IDs, paper-card links, experiment routes, and writing routes, use the generated [Evidence Index](/evidence/evidence-index) page.

## Claim Status Rules

* `verified`: supported by local project sources, paper cards, evidence matrix rows, or local experiment outputs.
* `inferred`: reasonable synthesis from verified evidence, but not directly proven for `SAH-BRI-Grasp`.
* `needs confirmation`: unresolved or dependent on missing experiments, hardware binding, or manual review.

## Candidate Claims

| Claim ID                       | Claim                                                                                          | Current Status | Evidence Backbone                                                              | Required Before Strong Writing                                            |
| ------------------------------ | ---------------------------------------------------------------------------------------------- | -------------- | ------------------------------------------------------------------------------ | ------------------------------------------------------------------------- |
| `CLAIM-PROJECT-SCOPE`          | SAH-BRI is the repository framework and SAH-BRI-Grasp is the current grasping system instance. | `verified`     | `README.md`; `framework/sah-bri.md`; `paper/manuscript.md`                     | Keep terminology consistent across docs and paper.                        |
| `CLAIM-CORE-HIGH-LEVEL-INTENT` | Non-invasive EEG is better scoped to high-level intent than dense low-level robot control.     | `inferred`     | `BCI-001` to `BCI-005`; `MI-005`; `MI-009`; `BRI-004`; synthesis mainline 1    | Exp2 and Exp3 evidence before performance claims.                         |
| `CLAIM-SSVEP-METHODS`          | SSVEP can serve as a scene target-selection channel.                                           | `inferred`     | `SSVEP-001`; `SSVEP-003`; `SSVEP-004`; `SSVEP-005`; `SSVEP-011` to `SSVEP-031` | Exp1 dynamic/frozen object-box results.                                   |
| `CLAIM-MI-MODE-CONTROL`        | MI can serve as mode control or intervention.                                                  | `inferred`     | `MI-002`; `MI-004` to `MI-009`; `MI-013` to `MI-032`; `BRI-004`                | Exp2 no-control, false activation, and latency results.                   |
| `CLAIM-YOLO-CANDIDATES`        | YOLO-style perception can generate a dynamic command space.                                    | `inferred`     | `YOLO-001` to `YOLO-010`; `system/pipeline.md`; `system/runtime-interfaces.md` | Candidate stability and target-binding results in Exp1/Exp3.              |
| `CLAIM-GRASP-BOUNDARY`         | Object detection alone is not grasp execution.                                                 | `inferred`     | `GRASP-001`; `GRASP-003`; `GRASP-005`; `GRASP-006`; `GRASP-007`                | Project grasp-pose and calibration implementation before physical claims. |
| `CLAIM-SHARED-AUTONOMY`        | Shared autonomy should arbitrate uncertain BCI intent and robot feasibility.                   | `inferred`     | `SA-001`; `SA-002`; `BRI-002`; `BRI-003`; `BRI-004`; `system/state-machine.md` | Exp3 baseline comparison and failure logs.                                |
| `CLAIM-CHANNEL-ABLATION`       | 16/12/8/6/3 channel ablation can test wearable EEG translation potential.                      | `inferred`     | SSVEP and MI low-channel comparators; EEGNet/FBCSP/benchmark evidence          | Exp4 offline ablation and online low-channel pilot.                       |
| `CLAIM-REHAB-PRODUCT-BOUNDARY` | Rehabilitation is a possible product direction, not a current clinical efficacy claim.         | `inferred`     | `MI-030`; `MI-031`; `MI-032` only as context                                   | Clinical protocol, ethics, safety case, and patient-centered outcomes.    |

## Blocked Output Claims

These are intentionally not promoted into `library/EVIDENCE_MATRIX.md` as active evidence rows because no local result evidence exists yet.

| Blocked Claim                                                                                                     | Status               | Required Evidence                                                                                         |
| ----------------------------------------------------------------------------------------------------------------- | -------------------- | --------------------------------------------------------------------------------------------------------- |
| SAH-BRI-Grasp improves task success, completion time, workload, grasp success, or low-channel online performance. | `needs confirmation` | Exp1-Exp4 outputs, statistics, exclusions, confidence intervals, workload records, and failure-mode logs. |
| SAH-BRI supports clinical rehabilitation benefit.                                                                 | `needs confirmation` | Separate clinical protocol, ethics review, safety case, patient cohort, and patient-centered outcomes.    |

## From Claim To Site Path

| Claim Need                 | Site Path                                                                   |
| -------------------------- | --------------------------------------------------------------------------- |
| Find current claim status  | `Evidence Matrix`; `Evidence Index`; `Claim Map`; `Gap Audit`               |
| Find supporting papers     | `Evidence Index`; `Synthesis Mainlines`; `Paper Library`; literature tracks |
| Find missing experiment    | `Open Questions`; `Experiment Roadmap`; Exp1-Exp4 pages                     |
| Write manuscript text      | `Paper Kit`; then cite evidence-backed paper IDs                            |
| Decide product positioning | `Product Directions`; keep product claims separate from experimental claims |

## Writing Boundary

The paper can currently make framework, motivation, design, and protocol claims. It cannot yet make strong system-performance claims because no local experiment data exist.