This illustration summarizes the structure of this work. We first collected illustrations of vibrotactile experiences from scientific literature (A), then developed and applied a novel taxonomy of illustration characteristics to code them (B). We furhter analyzed the codes with descriptive statistics and used them to identify potential shortcomings in several illustrations for which we provide more comprehensive alternatives (C). We encourage the readers to take a look at the paper to better grasp the illustrations we analyzed, as we provide examples with consent from authors or publishers.

The taxonomy is presented in the image above. It provides two primary categories of codes that indicate what the illustration is representing, and how it is conveying specific bits of information. When reviewing vibrotactile experience illustrations, the analyst can take this taxonomy and simply check each code that is being used by the illustration. While this tool may be difficult to fully grasp at first because of the complexity and sometimes abstraction of several codes, it becomes quite helpful to lead analyses of one's or others' work once mastered. This taxonomy builds on previous work and reuses codes from the following literature: Antoine et al. 2021, Buxton 1983, Culbertson et al. 2018, MacLean et al. 2022 and Bertin's Semiology of Graphics (1983).

We conducted a formative study to evaluate 1) whether the taxonomy is understandable and usable, and 2) how it supports the design of VTX illustrations. This study included participants with and without experience in haptic design and VTX to assess whether prior knowledge is required to make use of the taxonomy. They were provided with textual prompts describing VTX and asked to illustrate the experience. We used a within-subject design and captured textual data through semi-structured interviews (verbal transcripts). We recruited 11 participants from our institutes and research network with different experience levels in haptics and VTX (7 novices and 4 experienced). Results show that the taxonomy can support designers in reviewing their sketches and reflect on their choices, but not necessarily guide them. The taxonomy can “spark ideas” and “be helpful” to “know your options”, however, participants remarked it did not directly provide guidelines on how to create illustrations from scratch. This was indeed not the goal of this work as we aimed to identify the existence and use of atomic visual characteristics of illustrations, rather than analyzing how efficiently and why they convey specific information.