AUTHOR=Johnson-Glenberg Mina C. , Kosa Mehmet , O'Rourke Holly Patricia TITLE=STEM learning, science identity and immersivity: Giant screen films comparing 2D, 3D, and dome formats including a videogame assessment JOURNAL=Frontiers in Education VOLUME=7 YEAR=2023 URL=https://www.frontiersin.org/journals/education/articles/10.3389/feduc.2022.1096889 DOI=10.3389/feduc.2022.1096889 ISSN=2504-284X ABSTRACT=

Multiple assessments, including a novel videogame methodology, were used to understand how viewing a Giant Screen film in four different formats affected both science identity associated with liking science and desire to be a scientist, and learning gains on STEM content. Participants were 406 5th graders who were randomly assigned to watch the 45-min film called Amazon Adventure in either a (1) 2D Small Screen, (2) 2D Giant Flat Screen, (3) 3D Giant Flat Screen, or (4) curved Dome condition. The conditions increased in levels of immersivity as listed. In this 4 × 3 design, the first factor was viewing condition and the second was test time including pretests, posttests, and two-month followup on content knowledge tests relating to natural selection. The strongest pattern was seen at the followup time point where the three more immersive conditions (2D Giant Screen, 3D Giant Screen, and Dome) outperformed the Small Screen condition on the science identity measures of liking science and wanting to be a scientist when grown up. Additionally, two of the more immersive conditions performed significantly better at both posttest and followup on the videogame that assessed natural selection and mimicry knowledge. Significant gender effects at followup revealed that females in the two most immersive conditions (3D Flat and Dome) liked science significantly more and wanted to be a scientist more than males when grown. Interestingly, females also played the videogame significantly better than the males at both posttest and followup. These results suggest that increases in the immersivity of a film can have longer term effects on science identity, and those effects are stronger in females. Additionally, as more assessments are being designed in a gamified manner, designers/researchers should keep game-style (i.e., strategy games) and gender interactions in mind. It had not been predicted that the females would outperform the males on the STEM assessment videogame, but deduction games like the one in this study are probably capturing more than knowledge about natural selection, they may be a proxies for general intelligence or g. More research is needed on game type and gender effects during the school-age developmental span.