Three experiments were conducted to assess the effects of nonbizarre vs. bizarre pictorial elaboration on the paired-associate retention of noun pairs. Five and seven year old children served as subjects in the first two experiments and learned a list of common noun pairs by the study-test paired-associate method. Experiment 1 provided a between-subjects analysis of bizarre elaboration effects, while Experiment 2 provided a within-subjects analysis. The results of the two experiments indicated that both nonbizarre and bizarre pictorial elaboration faciliated performance relative to a standard presentation condition. A reliable difference between the two types of elaboration was not observed. Experiment 3 was designed to assess the effects of nonbizarre vs. bizarre pictorial elaboration on the seven day retention of noun pairs. The results of this final experiment indicated that the forgetting of pairs encoded under the different types of elaboration (nonbizarre vs. bizarre) was equivalent. (Author)