The main purpose of the present study was to determine whether exemplar training in symmetry relations would readily facilitate the transformation of function in accordance with symmetry, when subjects were not provided with explicit name training. The study also examined whether pretraining that was formally similar to the symmetry test, but did not reinforce symmetry relations, would have the same facilitative effect as exemplar training. Sixteen children, aged between 4 and 5 years, were employed across three experiments (i.e., 4 children each in Experiments 1 and 2, and 8 children in Experiment 3). In Experiment 1, subjects were trained in an action-object conditional discrimination using familiar actions and objects (e.g., when the experimenter waved, choosing a toy car was reinforced, and when the experimenter clapped, choosing a doll was reinforced). Subjects were then exposed to a test for derived object-action symmetry relations (e.g., experimenter presents toy car-*child waves and experimenter presents doll-*child claps). Across subsequent sessions, a multiple-baseline design was used to introduce exemplar training (i.e., explicit symmetry training) for those subjects who failed the symmetry test. Experiment 2 replicated Experiment 1, except that the trained and tested relations were reversed (i.e., train object-action, test action-object relations). Experiment 3 replicated Experiment 1, except that subjects were exposed to object-action pretraining. Across Experiments 1 and 2, none of the 8 subjects show derived objectaction (Experiment 1) or action-object (Experiment 2) symmetry until they received explicit symmetry training. Pretraining objectaction responding in Experiment 3 appeared to facilitate symmetry, but only for 4 of the 8 subjects. For the 4 subjects who failed , symmetry emerged following exposure to exemplar training. Overall, the data are consistent with Relational Frame Theory.