Spatiotemporal visualizations pose certain challenges for comprehension and reasoning processes. This specifically applies to casual users who want to gain an overview of the spatiotemporal origins of large cultural collections in a relatively short time. Based on a distributed cognition approach, we compare four visualization techniques (coordinated multiple views, color coding, animation and space-time cube representations), which offer different solutions to represent spatial and temporal information in an integrated fashion. We firstly assess the strengths and weaknesses of these techniques with regard to existing work. We then complement these findings with a close look at the reasoning processes of 18 casual users. In a mixed methods study, they explored a photo collection from the middle of the 20th century by means of all four spatiotemporal visualization techniques. We investigate and compare the effects of these techniques on processes of internalization, spatiotemporal reasoning and comprehension by analyzing the observed spatial, temporal and spatiotemporal insights and the degree of connectedness of these insights. Our results provide an in-depth look at how each visualization technique supports the construction of coherent internal representations and, together with data on cognitive load and subjective references, help to understand for which tasks and users they are suited.
Comparative Analysis of CNN-Based Spatiotemporal Reasoning in Videos
