We study the different explanations of the European Muon Collaboration effect. Although the theoretical models are very sensitive to input free-nucleon structure functions, we find that each has distinctive testable traits. Therefore, we examine different reactions to see which can distinguish the models. Deep-inelastic (anti)neutrino-nucleus scattering data are examined, but we find that the model differences to be exposed by this process are small. J/\ensuremath{\psi} production in lepton-nucleus collision is, in principle, capable of determining the nuclear gluon distribution, but there are too many difficulties to successfully separate the different models. We find that the Drell-Yan process is better suited than antineutrino scattering to determine the nuclear ocean distributions. These may be studied in an experimentally accessible kinematic region in which a quark from the projectile annihilates with a nuclear antiquark.