This study explored the effects of 2D- versus 3D-based media representations on the influence of the spatial visualization ability of undergraduate science majors. A pre-test/post-test comparison- group experiment was conducted with 23 participants involved in the study. Participating students were randomly assigned either to the interactive 3D media representation group (n = 13) or the con- ventional 2D media representation group (n = 10); learning materials in both groups deliver the same information to students, but employ different media representations. All the activities were per- formed in a self-paced, web-based instructional system. The results of ANCOVA analysis showed statistically insignificant difference between groups in terms of students' post-test scores on the spa- tial visualization ability test with the students' pre-test scores as the covariate. However, a medium effect size was observed in favor of the 3D group in terms of practical significance. As a pilot study with a small sample size aiming to probe the research direction of this problem, the result of medium- sized effect magnitude is likely to implicate that the discrepancy of different representational design on students' performance of spatial ability assessment is noteworthy. Future study of this nature appears to merit further replications and investigations.
Computers in Human Behavior - COMPUT HUM BEHAV , vol. 23, no. 4, pp. 1943-1957