Ratcliff, Van Zandt, and McKoon (1999, Psych. Rev.) claim that connectionist models fail to simulate many aspects of how individuals select one of two possible responses. Here, these claims are re-evaluated via computational and behavioral investigations of an extended version of the original numerosity judgment task. The results of the experiment indicate that some of the empirical effects that the models failed to capture do not generalize and were likely due to idiosyncratic aspects of the original methodology. The simulations show that a more biologically-plausible model captures the bulk of the new effects, including some trial-by-trial adaptive effects that are outside the scope of models tested against aggregate data, and emergent asymptotic stability that has previously required an explicit leak parameter.