Donghan Hu
Virginia Tech, Blacksburg, Virginia, United States
Effective self-tracking in working contexts empowers individuals to explore and reflect on past activities. Recordings of computer activities contain rich metadata that can offer valuable insight into users' previous tasks and endeavors. However, presenting a simple summary of time usage may not effectively engage users with data because it is not contextualized, and users may not understand what to do with the data. This work explores time-lapse videos as a visual-temporal medium to facilitate self-reflection among workers in productivity contexts. To explore this space, we conducted a four-week study (n=15) to investigate how a computer screen's history of states can help workers recall previous undertakings and gain comprehensive insights via self-reflection. Our results support that watching time-lapse videos can enhance self-reflection more effectively than traditional self-tracking tools by providing contextual clues about users' past activities. The experience with both traditional tools and time-lapse videos resulted in increased productivity. Additionally, time-lapse videos assist users in cultivating a positive understanding of their work. We discuss how multimodal cues, such as time-lapse videos, can complement personal informatics tools.
doi.org/10.1145/3613904.3642469
The ACM CHI Conference on Human Factors in Computing Systems (https://chi2024.acm.org/)