History-Dependent Mode Selection in a Driven Holographic Superfluid

We study finite-rate mode selection among competing Landau-unstable modes in a holographic superfluid. The selected mode is not generally the instantaneous fastest-growing one, while time-dependent linear evolution on the driven background reproduces the full dynamics. Protocols with identical present driving conditions can exhibit opposite modal ordering, revealing memory of the preceding drive. This memory is quantitatively captured by the accumulated difference between the QNM growth rates of the competing modes.