How do passive haptics encourage embodied learning compared to traditional keyboard interactions for online education? We integrated a 3D-printed burette into a virtual titration experiment to explore how passive haptics affected learning outcomes! 136 college students went through our learning experience in one of two conditions: 1) using keyboard arrows, or 2) physically turning a valve/stopcock on the 3D-printed burette. We found that students with higher prior knowledge preferred the 3D-printed burette, and students with lower prior knowledge preferred the keyboard. For educators designing mixed reality curriculum, it may benefit lower-prior knowledge students to begin with a more ubiquitous interface such as the keyboard before moving them to more multi-modal mixed reality devices.

The intersection of educational multimedia and 3D-printed learning objects holds much promise for STEM instruction. There is an opportunity to study how learners interact with 3D-printed materials that are tethered to interactive actions on a computer screen. Designers and creators of educational multimedia content need evidence-based design guidelines that balance the benefits of haptically-mediated platforms with the potential for distraction or cognitive overload.

One of the goals of this study is to understand how users’ actions and gestures that are afforded by a 3D-printed burette affect downstream learning and recall of educational content. This current study focused on titration. Learners were randomly assigned to either a control or experimental condition. In the control condition they pressed keyboard buttons to release virtual liquid from a virtual burette; in the experimental condition participants could physically turn a stopcock valve on a 3D-printed haptic burette to release the virtual liquid. The turning movement is gesturally congruent to what students do in real world, in-person chemistry labs. This mixed reality device should serve to prime and reify real world skills.

In both conditions, the further clockwise that the participants turn the virtual burette stopcock, the faster the virtual solution drops from the burette. Participants either controlled the virtual stopcock on the virtual burette by using 1) the keyboard-controlled burette or 2) the 3D-printed haptic burette. During the learning phase, participants in the control condition use the keyboard-controlled burette, and participants in the experimental condition use the 3D-printed haptic burette.

This study reveals some intriguing aptitude by treatment interactions based on a student’s prior science knowledge scores. We recommend instructors to take into account the students’ knowledge profiles and assign the lower prior knowledge students first to a more common device (one example is a keyboard) for scaffolding or a type of gradual exposure to the content. After those students demonstrate some science knowledge gains, they could then be switched to the more novel, real world and gesturally congruent device (e.g., the 3D-printed haptic burette). It may be the case that adding a haptic channel to the learning signal (which is usually only visual and auditory) may be overloading for lower prior knowledge students.

These types of studies help to inform educators and creators of online science courses on how to make decisions regarding the delivery of laboratory instruction courses. They can make more informed decisions about when and which types of students should be moved over to a mixed reality interface. This may inform when instructors mail out 3D-printed kits to students, or when they invite students to conduct real experiments in a campus lab.