It is a common assumption that Visual Analytics systems should enable users to make sense of the
data presented to them. Nevertheless, it is not yet entirely clear how this can be supported
appropriately. There are several theories developed in Cognitive Psychology and HCI which might be
adopted to explain such processes, but empirical research in Visual Analytics based on these
theories is sometimes not a straightforward process. This talk tries to clarify some issues related
to this problem. I will take three theories (graph comprehension, distributed cognition and
sensemaking) as examples to illustrate these issues.
The theory of distributed cognition, for example, takes the nature of the cognitive tool which
is used for sensemaking into account. This yields rich data on how the design of a visualizations
may influence the sensemaking and reasoning processes. On the other hand, it is sometimes difficult
to generalize these results because they are highly specific and only relate to the visualization
used. Sensemaking and graph comprehension theory, on the other hand, try to explain more general
reasoning processes and, therefore, yield more generalizable results.
In general, it can be argued that the theories have specific strengths and weaknesses, and it
depends on the research question under investigation which theory can best be used to explain the
results of a specific empirical investigation.
Habilitation (Univ.Doz.), Computer Science, Vienna University of Technology, focus
on: Human-Computer Interaction, 2002
Doctor of Philosophy (Dr.), Psychology, University of Vienna, focus on: Cognitive Psychology,
Master of Science (Mag.), Information Systems, University of Vienna, focus on:
Current position: Associate Professor (Vienna University of Technology)
http://igw.tuwien.ac.at/hci/ for more details.