For one year, 126 third through sixth grade elementary school pupils viewed an NHK year-long TV series on environmental education, "green earth" (Midori no chikyu). The pupils were permitted to expand their studies after viewing selected programs within the series. Pre- and post-tests were conducted with a two and a half month interval in order to evaluate the children's audiovisual literacy. Also, the children were asked after each program to write down their impressions and explain what they considered the producer's intentions to be. The main results of this study were: (1) It was possible to develop several test instruments for evaluating audiovisual literacy. These included a knowledge test for determining conceptual knowledge retention levels, a grouping test for evaluating knowledge systemization, an image-map for quantitatively and qualitatively evaluating images, a theme-comprehension test for interpreting the producer's aims, and a semantic differential technique for measuring emotional strength. (2) Factor analysis methods were used in order to determine the structure of the various facts of audiovisual literacy as measured by the above tests. As a result, each facet was categorized under one of two factors: cognition and image. (3) Two interesting patterns were located in regards to IQ and systematization of knowledge, one of the facts of audiovisual literacy. One pattern was that while there were almost no correlations found during the pre-test, a high degree of correlation was found during the post-test and individual differences for this facet of audiovisual literacy had increased. The other pattern was a high degree of correlation and large individual differences during the pre-test, but almost no correlation during the post-test. The first pattern was exhibited by third and fifth grade pupils while the latter was found in fourth and sixth graders. (4) Rather large differences in the content and form of the studies expanded by the third and fourth grade pupils versus those done by the fifth and sixth graders were seen. In other words, the studies of the two higher grade levels exhibited more varied and more advanced methods and content. In addition, they were more likely to form groups based upon separate theme projects and these pupils more often shared similar interests by adapting to the circumstances when forming their groups. In comparison, the third and fourth grade groups were more likely to be fixed in nature and the content of the studies were more likely to be simple and stereotyped.