The testing effect occurs when learners who are tested rather than relearning material perform better on a final test than those who relearn. Based on cognitive load theory, it was predicted that the testing effect may not be obtained when the material being learned is high in element interactivity. Three experiments investigated conditions of the phenomenon using school students aged 8 to 9 years learning to interpret and use a bus table. Results from Experiments 1 and 2 indicated a reverse testing effect on an immediate test. Experiment 3 was similar to Experiments 1 and 2 but had a 1-week delayed test which showed no significant effect. It was suggested that the failure to obtain a testing effect was determined by the high levels of element interactivity of the material that had to be learned.