Designing the Optimal Augmented Reality Experience

Augmented Reality (AR) is a technology that superimposes computer-generated images onto a user’s view of the real world allowing them to see the digital graphic in an approximation of what it would look like if it were actually there. While the AR features implemented on apps such as Instagram can be entertaining — such as making a tiger appear in your home — AR has found useful applications across a wide range of industries such as hospitality, healthcare, and manufacturing. It can be used to create virtual guest experiences, help physicians understand the functioning of a new medical device, simulate the production process on a factory floor, and more. Whether for entertainment or education, the development of these experiences is complex, and requires a disciplined approach based on best-practice techniques.

ESTABLISHING BEST PRACTICES FOR AR INTERFACE DESIGN

Best practices for interface design are crucial tools for designers and developers, who utilize these tools on a regular basis in order to guide interface design and optimize a user experience. While the guidelines, and subsequent techniques, for 2D interface design have been refined over many years thanks to the excellent work of the Nielsen-Norman Group (NNG) and others’ efforts, the guidelines for extended reality (ER) and augmented reality (AR) experiences have not.

In order to take a deep dive into AR design, we started with the Nielsen-Norman Group’s classic heuristics, which Oxford defines as “an approach to problem solving or self-discovery that employs a practical method,” and evaluated which heuristics were most applicable to AR and how best to apply them to AR interface design.

Of all of the research that’s been conducted into XR-specific heuristics, i.e, “Extended Reality” (XR) technologies including VR (Virtual Reality) and AR (augmented Reality), we found this article to be the most thorough and useful: Heuristic Evaluation of Virtual Reality Applications by Alistair Sutcliffe and Brian Gault. Sutcliffe and Gault used NNG’s heuristics as a foundation, then adapted, expanded, and applied them to the design of a virtual environment.

Sutcliffe and Gault used the below heuristics to efficiently evaluate the usability of two virtual reality applications, and we found them particularly useful in our development of a prototype for an AR retail shopping aid.

WHAT WE DISCOVERED

As we designed, built, and tested this prototype, we uncovered other best practices and opportunities for further study. For example, users unexpectedly expressed preferences for using the application in both portrait and landscape mode in testing. This desire for choices should be accounted for in AR application layout design, in general. We plan to conduct several studies of user behavior to uncover other AR best practices, including interactive element affordances, CX prototyping capabilities, and onboarding experiences. The results of these investigations will be covered in future posts.

In addition to general usability heuristics, we discovered a set of ergonomic guidelines that are very useful in determining the placement of AR elements in virtual space. The standard reach of a human arm, the angles of direct and peripheral vision, and the maximum distance of legibility all play roles in determining the best placement for interactable and visible objects in 3D virtual spaces. We have included a handy cheat-sheet below for reference and we will be building an AR-viewable reference in our next post to illustrate these concepts in their realized form.

VIRTUAL ENVIRONMENT ERGONOMIC LAYOUT DIAGRAMS

The diagrams above illustrate the ideal placement of interactable and readable elements in a virtual environment based on standard ergonomic measurements of humans. A user should have around 50 centimeters of “personal space” to avoid feeling cramped. Our normal vision forms a 60-degree cone from our eyes and our peripheral vision extends to 120 degrees.

Objects intended for immediate interaction should live within the 60-degree cone and within arm’s reach at around 70 centimeters. Objects indented for secondary interaction should live further out within that same cone, but no further than 20 meters. At longer distances, text gets difficult to read and objects hard to recognize. Finally, objects intended to be hidden and discovered can live behind the user or in their peripheral vision.

Below is a checklist to help you achieve good usability practices for any extended-reality application. As always, no set of rules replaces real testing, so allow ample time for that in your project’s timeline.