Download Reducing Cognitive Load Using Uncertainty Visualization

Reducing Cognitive Load Using Adaptive Uncertainty Visualization
Gregory Block
Advisor: Dr. Maxine Cohen
Problem Statement
Research Objective
Real-world environments are plagued with uncertainties, such as
faulty sensors, and sporadic network connectivity often caused by
environmental factors, including bad weather, and unplanned
intrusions (Girardin & Nova, 2005). These conditions can adversely
affects the decision-making process, and even add to the user’s
cognitive load (Zuk & Carpendale, 2006).
Applications with high levels of certainty can positively affect user
impressions by displaying the uncertainty (Lim & Dey, 2011).
However, adding uncertainty visualizations to a crowded visual
canvas can adversely affect a user’s cognitive load (Antifakos,
Schwaninger, & Schiele, 2004).
The literature has not provided a proven theory for effective
uncertainty visualization (Lapinski, 2009) and the effects of
uncertainty visualization on reasoning during periods of high
cognitive load.
Identify how adaptive uncertainty visualization can decrease cognitive load
arising from uncertainty more than visualization increases cognitive load
arising from complex user interfaces. By mitigating the increase in cognitive
load due to uncertainty, uncertainty visualization techniques can reduce
the user’s overall cognitive load.
uncertainty visualization…
The Five-Day Track Forecast Cone, a sight familiar to most Floridians
(National Weather Service)
Reduces anxiety related
to uncertainty
Research Questions
Research Methodology
1. Does adaptive uncertainty visualization improve the
system operator’s level of performance in completing
assigned tasks? Performance will be measured by reaction
time, and error rate.
2. Does adaptive uncertainty visualization reduce the system
operator’s mental load? Mental load will be measured by
the operator’s performance on the secondary task.
3. Does adaptive uncertainty visualization reduce the system
operator’s level of mental effort? Mental effort will be
measured by a subjective survey.
4. Can the adaptive uncertainty visualization be calibrated to
reduce cognitive load that stems from uncertainty without
increasing the system operator’s overall cognitive load?
•
•
•
•
•
Increases anxiety related
to cognitive load
Design a computer simulation that reproduces the effects of
uncertainty and high workload (Garrabrants, 2009)
Conduct studies using two conditions: with and without
uncertainty visualization (the independent variable)
Study sample size using 24 knowledge workers (12 male and
12 female)
Quantitatively measure the subject’s response time and error
rate using primary and secondary tasks (de Jong, 2009)
Qualitatively measure user satisfaction with a questionnaire
Computer simulation illustrating the subject’s primary task
Reference List
Project Status
1. Dissertation committee formed: Dr. Maxine Cohen, Dr.
Sumitra Mukherjee, Dr. William Garrabrants
2. Dissertation research proposal approved
3. Currently piloting software simulation
Pilot Findings
• Reduced simulation from four monitors to one due to
subject reports of feeling overwhelmed by four monitors
• Added an animated “busy” indicator during simulated
outages to retain the subject’s attention
• A training round is useful for immersing the subject in the
simulation
• Simulation should have high fidelity (closely matches a realworld situation) to increase subject’s buy-in
Simulation secondary tasks
Distractors that
contribute to
cognitive load
Uncertainty visualization
demonstrating the probability
region
Antifakos, S., Schwaninger, A., & Schiele, B. (2004). Evaluating the Effects of Displaying
Uncertainty in Context-Aware Applications. Proceedings of the 6th International
Conference on Ubiquitous Computing (pp. 54-69). Nottingham, UK: Springer-Verlag.
de Jong, T. (2009). Cognitive Load Theory, Educational Research, and Instructional Design:
Some Food for Thought. Instructional Science, 38(2), 105-134.
Garrabrants, W. M. (2009). Reducing Cognitive Load Using Hypervariate Display. Doctoral
Dissertation. Nova Southeastern University.
Girardin, F., & Nova, N. (2005). Getting Real with Ubiquitous Computing: the Impact of
Discrepancies on Collaboration. International Journal on Human-Computer Interaction,
1(1), 60-64.
Lapinski, A.-L. S. (2009). A Strategy for Uncertainty Visualization Design. Defence Research
and Development Atlantic Dartmouth (Canada).
Lim, B. Y., & Dey, A. K. (2009). Assessing Demand for Intelligibility in Context-Aware
Applications. Proceedings of the 11th International Conference on Ubiquitous Computing
(pp. 195-204). New York: ACM.
Zuk, T., & Carpendale, S. (2006). Theoretical Analysis of Uncertainty Visualizations. Proc.
SPIE & IS&TConf. Electronic Imaging,Vol. 6060: Visualization and Data Analysis.