The problem of track-to-track association–a prerequisite for the fusion of tracks–has been considered in the literature for tracks described by kinematic states and, more recently, has been generalized to include additional (continuous valued) feature and (discrete valued) attribute variables which pertain to those tracks. These approaches allow the search for the maximum likelihood (ML) or maximum a posteriori (MAP) association. However, while for kinematic variables there is a “gating” procedure based on a Gaussian distribution–which corresponds to a Neyman-Pearson test of “common origin” (actually, “same kinematic state”) with selectable power–there is no simple counterpart of this for attributes. The sufficient statistic for the optimal association test (in the Neyman-Pearson sense) based on discrete-valued target classification information observables (attributes) is derived and its relationship with the class probability vector is discussed. Based on this, “attribute gates” are presented, which allow a Neyman-Pearson test for “same class” with the desired power.