Previous studies on naming have presented the object and its name simultaneously during both training and testing, and thus the training component may establish a transformation of function directly between the object and the name. Successful tests for listener naming may thus not require the emergence of a novel (entailed) transformation of function. The current study aimed to control for this possibility by presenting the object and the name sequentially and nonsimultaneously. Eight typically developing toddlers participated in the current study. During name training, objects and names were presented nonsimultaneously, and all participants failed to emit listener-naming responses during the first test session. Subsequently, 4 participants received multiple exemplar training, which led to improvements in listener naming for all 4; and speaker naming for only 1 participant. As a control condition, the remaining 4 participants were tested repeatedly, without multiple exemplar training, and did not show any consistent improvements in their listener or speaker performances. Multiple exemplar training thus appeared to be effective in establishing generalized listener responses, which involved generating entailed transformation of functions. The strategy of using nonsimultaneous stimulus presentations could allow for greater precision in identifying the behavioral processes involved in listener-naming.