Is there an optimum of realism in computer-generated instructional visualizations?

Abstract

Realistic visualizations are abundantly used in digital education. However, the use of realism is still thought to risk a cognitive overload due to excessive details. Moreover, it is still not precisely known whether there is an optimal level of realism that benefits learners the most. In two experimental studies, different versions of anatomical visualizations were compared regarding their effects on retention performance and the subjective cognitive load experienced during learning. In Experiment 1 (n = 73), four visualizations with minor variations in the geometry and shading of the model featured in the visualizations were used. Although neither the level of detail in the geometry nor the realism of the shading resulted in significant differences, a detailed model with simplified shading elicited the highest retention scores descriptively. In Experiment 2 (n = 156), a schematic visualization was compared with an “idealized” model featuring only simplified shading and a highly realistic rendering. There most realistic version elicited the highest retention scores, but also the highest cognitive load ratings. Taken together, the results suggest that the optimal level of realism might lie on the more realistic end of the spectrum for learning tasks focused on the memorization of shapes that are assessed using image-based tests.

Realistic details impact learners independently of split-attention effects

“…However, Experiment 2 neither supports an amplifying (H 1a ) nor a compensatory (H 1b ) effect of realism. In contrast to the previous experiments (e.g., Skulmowski 2022a ; Skulmowski and Rey 2021 ), a higher level of realism did not help, but diminished retention performance. The retention results of Experiments 1 and 2 do not support the claim that realism is able to worsen other visual demands that may exist in a visualization.…”

“…While Experiment 1 features schematic test visualizations, Experiment 2 contains realistic rendering in the testing stage. The choice of non-biased test visualizations has been discussed in the literature (e.g., Skulmowski 2022a ; Skulmowski and Rey 2018 ). Based on a previous study, learning with realistic visualizations can lead to particularly high retention scores if the test visualizations are equally realistic, while the use of schematic visualizations in the testing stage does not appear to favor those who learning using schematic representations (Skulmowski and Rey 2021 ).…”

“…For instance, a study on anatomy learning found that learning with realistic visualizations mainly has positive effects if the learning test makes use of the experience with the realistic representation, such as if the images used in a learning tests were also prepared in a realistic style (Skulmowski and Rey 2021 ). Moreover, two studies have revealed that higher levels of realism can result in better retention performance (or better performance for parts of a visualization that were rendered realistically), but at the same time increase subjective cognitive load (Skulmowski 2022a , b ; Skulmowski and Rey 2020 ). However, the desire to explore objects has been found to increase with higher levels of perceived realism (Höst et al 2022 ), providing another potential advantage of realistic visuals for learning.…”

Guidelines for Choosing Cognitive Load Measures in Perceptually Rich Environments

“…Recently, a strict understanding of these ideas has come under fire in response to several results from digital learning involving perceptually rich media. In a number of recent studies, a higher measured (or assumed) cognitive load has actually resulted in higher levels of performance (e.g., Seufert, Wagner, & Westphal, 2017; Skulmowski, 2022; Skulmowski & Rey, 2020b; Skulmowski & Rey, 2021; Tugtekin & Odabasi, 2022). This seemingly contradicting aspect has been summarized by Skulmowski and Xu (2022) as an inherent principle of (perceptually rich) digital media and virtual environments.…”

“…However, the studies summarized in Table 2 support the finding of Ayres et al (2021) that single‐item surveys are still holding their ground in comparison to multifaceted surveys. The multifaceted cognitive load surveys revealed an important insight concerning instructional design: In three recent experiments, a higher level of extraneous cognitive load was not associated with lower, but rather higher learning performance (Skulmowski, 2022; Skulmowski & Rey, 2020b; Tugtekin & Odabasi, 2022).…”