How do you separate a real rate-differential move from session noise in majors?

When a currency pair moves sharply, it helps to ask whether the move is driven by the underlying policy-rate and yield-curve story or by the session it happened in. Two-year yield differentials between the relevant central banks tend to be the cleaner signal over days and weeks, while the same pair can gap or whipsaw during thin Asian hours or around the daily rollover with little change in fundamentals. A practical habit is to check whether the move persists once liquidity normalises, for example after the London and New York overlap. If the yield differential has barely shifted but price has, that points toward positioning, stop clustering, or quoted-spread effects rather than a change in rate expectations. Spreads themselves can widen at those times, so a move that looks large on a mid-price chart may be smaller once you account for the real cost of crossing the spread. Where I'm less sure is how much weight to give the front end versus the belly of the curve when central-bank guidance is mixed. Which indicator do you rely on to decide that a rate-driven move is genuine rather than session-driven?