Children’s tendency to associate novel words with novel refer-ents has been taken to reflect a bias toward mutual exclusivity.This tendency may be advantageous both as (1) an ad-hoc ref-erent selection heuristic to single out referents lacking a labeland as (2) an organizing principle of lexical acquisition. Thispaper investigates under which circumstances cross-situationalneural models can come to exhibit analogous behavior to chil-dren, focusing on these two possibilities and their interaction.To this end, we evaluate neural networks’ on both symbolicdata and, as a first, on large-scale image data. We find thatconstraints in both learning and selection can foster mutual ex-clusivity, as long as they put words in competition for lexi-cal meaning. For computational models, these findings clarifythe role of available options for better performance in taskswhere mutual exclusivity is advantageous. For cognitive re-search, they highlight latent interactions between word learn-ing, referent selection mechanisms, and the structure of stimuliof varying complexity: symbolic and visual.