The core factual assertion — that the first $400 million of the $1.7 billion Iran settlement was paid in January 2016, on the same day as JCPOA Implementation Day and the American prisoner release — is substantially confirmed by multiple Tier 1 congressional records and independent Tier 4-5 sources. However, Psaki's claim contains two identifiable inaccuracies: (1) She says 'early January,' but every authoritative source confirms the disbursement occurred on January 17, 2016 — conventionally mid-month, not early January. Congressional testimony under oath from a Treasury official specifies 'January 17th' explicitly. (2) She says 'four Americans,' but congressional testimony and Brookings document five Americans were released (four in the formal swap, one — Matthew Trevithick — separately on the same day), though some contemporaneous media reporting also used 'four' when referring to the formal swap. The gate1 verdict nominated MISLEADING via FRAMING, reasoning that the 'early January' phrasing constitutes a distortive framing of an accurate underlying event. Gate 2 applies the MISLEADING materiality test: the distortion must 'change the directional implication for a reasonable viewer.' The temporal qualifier 'early January' versus 'mid-January' (January 17) is a factual inaccuracy — it is simply wrong — rather than a framing device that creates a false impression about the nature or significance of the event. A reasonable viewer's understanding of the substance of the claim (settlement payment concurrent with nuclear deal implementation and prisoner release) is not directionally changed by whether the payment occurred on January 3 or January 17. Per the methodology's boundary test, MISLEADING requires that every specific fact in the claim be confirmable and that the distortion mechanism operates through omission, cherry-picking, distortive framing, or outdated presentation of technically accurate facts — not through a straightforwardly incorrect factual predicate. 'Early January' is not a framing technique applied to an accurate fact; it is an inaccurate statement of when in January the payment occurred. This places the verdict at MOSTLY TRUE: the core directional assertion is substantially correct, and the inaccuracies (temporal qualifier, prisoner count) do not reverse the claim's meaning, but they are identifiable and confirmable errors (per Section 3.2, MOSTLY TRUE requires the core assertion to be substantially correct with identifiable inaccuracies that do not reverse directional meaning). No correction was issued.

Methodology note: The FRAMING trigger is structurally constrained to cases where the underlying facts are technically accurate but presented in a distortive way — the canonical example being accurate data presented with misleading comparisons or denominator manipulation. Where the claimed fact is itself wrong (here: the date 'early January' is factually incorrect), the appropriate verdict category is MOSTLY_TRUE or FALSE, not MISLEADING, per the protocol's own instruction that 'if a specific fact is wrong, the verdict is MOSTLY TRUE or FALSE, not MISLEADING.' This boundary is worth documenting for pipeline calibration: Gate 1 applied FRAMING to what is more precisely characterized as a factual inaccuracy in a temporal qualifier.