The present study was conducted in order to evaluate pictorial elaboration effects in children's incidental paired-associate memory. The design of the experiment consisted of a 2 x 2 x 2 x 5 factorial with (1) grade level (kindergarten vs. second), (2) pictorial presentation (standard vs. elaborated), (3) list (two 14-pair lists of common noun referents), and (4) instruction (a memory control and four different incidental memory conditions). One hundred and twenty children at each grade level were tested individually by the study-test paired-associate method. The results indicate that pictorial elaboration facilitated paired-associate performance relative to standard presentation; this outcome was not influenced by either the factors of age or of instruction. The results were interpreted in terms of an automatic encoding hypothesis, which suggests that the intention to relationally encode pair members is not necessary for elaborative coding. The mere inspection of referents depicted under pictorial elaboration is sufficient for the induction of elaborative coding, hence the facilitation of paired-associate performance. (Author)