Simulations and virtual or remote laboratories are increasingly used in schools. The extent to which individual experimental skills can be acquired when experimenting in digital applications is, however, questionable. This paper focuses on finding multimedia features for digital experiments to support the transfer of measured values from the laboratory system to a diagram. Beside physical considerations, spatial translation processes could be crucial for a successful assignment. Therefore, the influence of the subjects' spatial ability is examined. Using a pretest post-test design (N ¼ 119), the effects of training with supportive animation (animation group) and training with an interactive task and feedback (interactive group) were tested. The results of both groups were each compared to those of a reference group. Eye tracking data were recorded during training to investigate the origin of different training effects. Hence, fixations and saccades during training were analyzed. For the investigation of the distribution of the saccadic movements, polar diagrams were used in combination with estimated probability density functions. The results show that the score in the pretest is correlated to the score achieved in the card rotation test, which measures the spatial rotation skills of the subjects. Further, the subjects in the interactive group benefited from the training more than the subjects in the reference group did. There were no significant differences in the effect of the training between the animation group and the reference group. Eye tracking data reveal that the training in the interactive group caused the most comparative eye movements between the laboratory system and the diagram. The training in the animation group led to the highest visual attention; however, subjects in this group concentrated on the dynamic elements. These results indicate that especially students with weak spatial skills need additional support when transferring measured values from the laboratory system to the diagram. This assignment can be practiced in computerbased experiments, in particular with an interactive training task and feedback. Additionally, the analysis showed that the training is equally suitable for learners with different spatial abilities. A corresponding task was implemented into a virtual laboratory.