Exploring Anachronistic Icons in Graphical User Interfaces — A Multi-Part Study Investigating Young Adults’ Perceptions and creativity
The graphical user interface (GUI) is one of the defining features of modern computing. Icons are one of the key elements of these graphical user interfaces — symbols used to represent functions and actions. As technology has advanced, many of the objects that these icons represent have become anachronistic — objects that are no longer used in our daily lives.
In this blog post, I talk about a study I conduced to explore the disconnect between the anachronistic objects represented by some of today’s interface icons and young adults’ perceptions of the objects themselves. To do this, I conducted a multi-part study to understand how well young adults know the objects in some of todays icons, what kind of icons they would come up with to replace outdated ones, and to assess how identifiable that set of icons elicited from young adults would be.
I recruited 30 young adult technology users ages 18–22. In the first part of the study, I presented them with 39 descriptions of computing functions that currently have an icon representing a plausibly anachronistic object. I asked them to sketch (and describe) icons that would trigger these functions. I elicited a total of 3,590 icons from the 30 participants, or an average of about 3 icon proposals per function from each participant. I then clustered similar icons to arrive at a set of 39 participant-generated icons.
In the second part of the two-part study, I conducted an identification study with the same 30 participants based on my end-user identification method—which I created as part of Ph.D. Thesis. In this study, I presented the participants with cards, each showing one plausibly anachronistic icon, and asked them to identify the computer function that the icon would trigger when clicked. I also asked the participants how familiar they were with the represented real-world objects themselves.
I conducted open-coding analysis on all 3,590 icons that I collected in the elicitation study and formulated a taxonomy of computing iconography. I found that almost half of the 3,590 participant-generated icons were of new concepts, while the other half were drawings of existing icons. When assembling the participant-generated set of icons from the first part of the study, I found that 17 out the 39 icons remained anachronistic, and 22 icons were of new concepts.
In the identification study, I found that only 16 of the physical objects were used by all 30 participants. None of the other 22 were used by all. 31 of the 39 anachronistic icons were identified correctly by the young adult participants. Of the 8 incorrectly identified icons, 5 were new concept icons. These 5 icons were identified correctly by a second study with 60 online participants. 34 of the 39 user-generated icons were identified correctly. 3 of the 5 incorrectly identified icons were new concept icons and 2 were anachronistic.
As a result of the three-part study (in-lab elicitation and identification, and an online identification study), I identified a total of 39 user-generated icons, which can be divided into three categories: Anachronistic by Elicitation, Anachronistic by Identification, and New Concept Icons. Of these, 20 were purely New Concept Icons, and 19 were anachronistic — 15 Anachronistic by Elicitation and four Anachronistic by Identification.
This finding is significant, as it highlights the importance of participant input in the icon development and design process. The study revealed that young participants were unfamiliar with most of the objects represented in today’s interfaces, but also proposed new icons that can better represent concepts. Additionally, the results of the online identification study suggest that the new concepts proposed by the participants are easy to recognize and understand.
Of course this was not a solo effort. I did this work as part of my Ph.D. dissertation and had an incredible team of collaborators. If you’re interested in reading the full published journal article, head over here.
