PhD Dissertation Proposal Defense: Zachary While, Toward Broadening Data Visualization Design to People in Late Adulthood
Content
Speaker
Abstract
Data visualizations are becoming increasingly ubiquitous in daily life, presenting data related to areas such as weather, health, and exercise. However, much of our existing empirical knowledge of visualization design has been built using samples of convenience made of younger participants (e.g., age 18-30). This leads to a knowledge gap regarding how visualization design may differ for people in the late adulthood stage (PLA), often characterized as starting around age 60 or 65. Furthermore, the global population is predicted to experience an overall demographic shift toward older age, with the number of PLA (ages 65+) expected to outnumber the number of children (ages 0-14) sometime between 2050 and 2075. Thus, it is of great importance and relevance to the visualization community to understand how various changes associated with aging (e.g., changes in perception) can be accounted for and incorporated into the current visualization design. My dissertation work lays the groundwork for this new subfield of visualization research called GerontoVis, establishing baselines for the visualization performance of PLA and bringing more attention toward this underrepresented population.
First, I establish the subfield of GerontoVis, arguing for the importance of expanding visualization research to include PLA and summarizing existing work to stimulate future work in the area. I further discuss how physiological changes due to aging may affect visualization task performance as well as how adjacent bodies of work in user interface design for PLA and visualization accessibility are, on their own, insufficient for supporting PLA due to the heterogeneous nature of PLA and the wide variety in how changes due to aging manifest.
Second, I conduct a conceptual replication study on the perceptual speed of PLA using glanceable visualizations on smartwatches. This type of visualization aims to concisely depict information so that it can be viewed and understood in a very short amount of time (i.e., at a glance). We replicate an existing study with a set of 24 participants aged 65 and older, tasking them with comparing two elements of a data visualization (bar, donut, or radial) for varying amounts of data (7, 12, or 24 data points) on a smartwatch. We find that differences in performance become stronger as age increases, with PLA ages 75+ requiring more time to accurately complete the comparison task compared to those ages 65-74.
Third, I carry out a study evaluating the impact of contrast polarity design (i.e., using dark mode or light mode) on the visualization performance of both younger adults (YA) as well as PLA (age 60+). Participants are shown and asked questions about different types of visualizations (bar, line, scatterplot, and pie), with each either depicted using positive contrast (i.e., dark foreground objects on a light background) or negative contrast (i.e., light foreground objects on dark backgrounds). We find that the impact of contrast polarity varied greatly across participants in each age group, with a roughly even distribution of participants having better performance (higher task accuracy and faster response times) for each contrast polarity. Additionally, participants’ preferred contrast polarity did not always match the one with which they experienced their best performance. This provides evidence that the impact of contrast polarity in the context of data visualization does not noticeably change with age.
My ongoing and future work consist of the following: (1) Performing a larger empirical study covering the impact of both visualization choice and user task on performance (accuracy and time), comparing results between YA and PLA. This can provide greater insights into how aging may affect performance as well as how those differences may be distributed at the individual level. (2) Analyzing the experiences of PLA using embedded information displays (EID) that communicate information relevant to home devices (e.g., microwaves, stovetops, and dishwashers) using a variety of data representations such as lights and LED screens with numbers and icons. By conducting interviews with a qualitative analysis, we can further understand design choices that can help and hinder PLA with daily device usage and can lead to devices that more effectively communicate information relevant to device operation. Each of these future avenues will further expand our understanding of visualization design for PLA through a wider variety of contexts, giving future researchers a stronger starting point for expanding this area of work.