Title:  Sketching Cognitive Models of Urban Spaces for the Visual Analysis of Spatial Data

Alex Godwin
Date: Thursday, January 12, 2017
Time: 10am to 12pm (EST)
Location: TSRB 223

Committee
--------------------
Dr. John Stasko (Advisor), School of Interactive Computing, Georgia Institute of Technology
Dr. Alex Endert, School of Interactive Computing, Georgia Institute of Technology
Dr. Keith Edwards, School of Interactive Computing, Georgia Institute of Technology
Dr. Christopher Le Dantec, School of Literature, Media, and Communication, Georgia Institute of Technology
Dr. Jason Dykes, School of Computer Science, City, University of London

Abstract
--------------------
Planning new policies in a city is a complex process involving competing interests from diverse populations. Geographic Information Science (GIS) toolkits are often used to support this process through analysis of relevant spatial data. GIS systems can be expensive and challenging to use, however, creating a prohibitive barrier that makes it difficult for non-experts to contribute their understanding of data to discussions of policy. Participatory workshops often incorporate sketch-based activities to elicit feedback from community members, but the paper artifacts created during these workshops remain disconnected from the rich quantitative data available. Sketch-based cognitive models can provide an external cognition aid that allows community members to confront and interpret their qualitative understanding of an urban environment during quantitative analysis of spatial data. These models encode the aspects of an area that a person considers core to the nature of that place, incorporates the significant elements of the urban environment, and reflects the beliefs of the individual who created the model. Visualizations created from these models can then serve as a boundary object for cooperative discussion of spatial data tied to a physical place.

In this talk, I will discuss previous and proposed research to incorporate quantitative data and cognitive models of the city. I will discuss my previous research on the visualization of public safety data in large cities, and sketch-based interaction methods for police officers and community members to analyze that data. I will discuss my plans for conducting a series of laboratory studies to evaluate the effectiveness of competing approaches for sketch-based interaction. I will also discuss my plans for conducting multi-site workshops for community participation in the design of cognitive model sketching methods, and the elicitation of feedback for the analysis of neighborhood models created by community residents and outsiders. Finally, I will describe how the results of the previous studies can inform the design and development of a web-based cognitive model sketching platform, which will be evaluated through deployment in neighborhood planning unit meetings in the city of Atlanta.