Conflict: Valence Vectors vs μ/κ Stability Overlays

Conflicting Claim IDs

Q-008
MECH-035
MECH-048
MECH-055

Verbatim Excerpts (with preserved sources)

From docs/architecture/control_plane.md:

“Should REE retain a dedicated valence vector … or can μ/κ stability overlays subsume the functional role of valence?”

From docs/architecture/control_plane.md:

“REE should keep three affect-related control axes distinct.”

Why They Conflict (or What Would Reconcile Them)

Q-008 asks whether valence can be absorbed by μ/κ overlays, while MECH-055 currently requires separation between hedonic stability overlays and valence appraisal. The conflict is whether that separation is architectural necessity or temporary scaffolding pending better calibration evidence.

Reconciliation Question

Is valence appraisal required as an independent axis for ranking and replay, or can a μ/κ-based regime model recover those functions without loss of behavioral discrimination?

Status

Resolved on 2026-02-18.

Resolution summary:

Valence remains a dedicated affective appraisal stream.
μ/κ overlays remain stability/commitment modulators and do not replace valence ranking semantics.
Remaining work is calibration and orthogonality testing, not axis collapse.

Resolution note:

docs/conflicts/resolutions/2026-02-18_valence-vs-mu-kappa.md