Research

The SEEE Tripartite Model

I am deeply interested in how humans perceive and learn about their environment. Especially in how information systems may be used to help mediate communication, interpretation, understanding, learning, decision support, action selection and creativity. This interest led me to investigate fields of human-computer interaction and then scientific visualization, as humans are excellent pattern recognition creatures, we can be most efficient with interactions in our environment when our information systems are designed with the human factors at the forefront.

The main direction of my research is theoretical and firmly framed in the domain of information science. Investigation is into 3D user-interfaces and the interplay of virtual environments, simulations, content, the mental model of the user and the search, navigation, augmentation, and annotation possibilities of the user-interface. Quantitative and causal models are used to explain the role that ecological context has in the dynamic interaction with cognitive models, and is at the core of this research program. This is an aggressive research program that requires collaboration between computer scientists, psychologists, educators and scientists. The research aim is to use qualitative data fitting techniques to investigate the design parameter interaction with mental models and thus evaluate the SEEE Tripartite Model. It is defined as a tripartite model, as it is believed to be a Markov Model.

Simulated Ecological Environments for Education (SEEE) Model, (2008-Present).

An investigation into the empirical models with the data generated by the simulations to understand the causal models and the relationships between the design factors of complex user interfaces in these virtual environments on human thought and action. 

The user interface (UI) is designed to enhance the perception of signals (various redundancy-gain techniques such as highlighting and scientific visualization) and to intentionally capture the user’s attention. Very much like an art student being trained to actively and consciously observe, see, and analyze the visual object, then to re-create it as artifacts, check it for error, and iteratively reduce the error, the user interface should support the user in learning to see more actively and consciously the details in the virtual environment (VE). The user interface can use various abstraction techniques to highlight important information, direct attention, and reduce cognitive overload.

Changes in the one's knowledge (Δ Knowledge) are represented in the framework as a novice’s ontology of the domain, a subset of expert’s, which, after interacting with the simulation (VE) and the user interface (UI), may iteratively expand outwards and towards the experts’ ontology of the domain. The experts’ knowledge is representative of both the declarative knowledge of the domain ontology, and procedural knowledge as the algorithms (rules) and the heuristics of the domain. Changes in the virtual environment result in new visual signals for the child to perceive. New signals present opportunities for the child to either recognize (accessible in memory) or not recognize (not in memory); that is, it is the opportunity to inquire. The active, chosen act of inquiry, based on the child’s understanding that they do not know, is a user-initiated action or a set of actions via the user-interface, requesting semantic information about that visual signal. This is active user-initiated inquiry, which is different from a passive reception of new information about an unrecognized visual signal, and is a very important distinction for this research. Since we only perceive what we can see and know to look for, we need to recognize unknown signals and seek to know them.

Figure 6., from my disserataion: Harrington, Maria, C. R. (2008). Simulated Ecological Environments for Education (SEEE): A Tripartite Model Framework of HCI Design Parameters for Situational Learning in Virtual Environments. Dissertation Abstracts International. July 17,2008. University of Pittsburgh, Pittsburgh, PA.

Current Research:

2012-2013: Investigation and Evaluation of Virtual Worlds, 3D Environments, and Simulations for Alternative Learning, Collaboration, Co-creating, and Decision Support Formats: The Virtual SRU Campus Modeled in Google Earth and Second Life for a Course on Sustainable Entrepreneurship.

ABSTRACT:
The purpose of this project is to learn how to best build, deploy and generalize the use of 3D virtual worlds and simulations, specifically Google Earth and Second Life, for teaching and learning across PASSHE. Virtual environments may be used for communication, collaboration, decision support, creativity and educational assessment. Our proposed project is to learn and test such tools. We plan to embed the curriculum of a class on sustainable entrepreneurship. These tools extend traditional teaching methods.  They will be used to express future realistic scenarios in context, for team-based decision support tasks of current-state simulations and multiple future-state scenarios and are expected to add substantially to the educational experience. Just as a spreadsheet may express multiple financial scenarios; our project is envisioned to allow multiple scenario co-creations and investigations of large complex physical and logical systems, thus pushing the boundaries of 21st century skills that go beyond Desire2Learn (D2L) and into the high-fidelity training simulations.

The Virtual Trillium Trail, (2003-2008). Project (PDF):

An educational simulation of elementary school environmental science curriculum and real world field trips, it was an investigation into the design, construction, testing, and evaluation of the child-computer-environment interface. The research was focused on the ethnographic design and empirical investigation into the design parameters that affect learning, activity, and emotional outcomes, also resulting in discovering gender based differences.

Aligned with Pennsylvania educational standards and augmented with an activity study of an expert naturalist guide from the Audubon Society of Western Pennsylvania, it is a comprehensive learning simulation which showed over 50% transfer effects from real to virtual and 100% again from virtual to real, and showed close to 40% learning gains as measured between a pretest and posttest in one hour of free play for the high fidelity and high navigation virtual condition, alone. Strong empirical evidence to use the real and the virtual together to maximize total learning, estimated at 190%, is the outcome of this study.

This research will contribute to the development of search and navigation tools that support children in independent exploration and intrinsic, deep learning and acts of creation. The project has collaborated with Ms. Gabi Hughes, Environmental Educator who coordinates the Fox Chapel Area School District environmental program, and other Naturalists from the Audubon Society of Western Pennsylvania, headquartered at Beechwood Farms Nature Reserve and third, fourth and fifth grade students, and with Professor Susan Kalisz and Senior Research Associate Jessica Dunn, Department of Biological Sciences, at the University of Pittsburgh. It is a unique data simulation and visualization of Dr. Kalisz’s biological plot study data, unlike any other virtual world for education. Furthermore, the location of the Trillium Trail Wildflower Reserve is the site for many science and ecology elementary school field-trips, created and lead by the Audubon Society of Western Pennsylvania. The combination of the biological data (from 1994 to the present), and the proven educational approaches (from 1968, Beulah Frey, Ruth Scott and Ruth Boyles) offered an ideal and successful real world learning experience to model and to base a computer simulation for research on the usability of virtual worlds and to better understand patterns of intrinsic learning found in informal settings.

VirtualTrilliumTrailStudyOne2007RealVirtualHarrington

>> Read the Spring 2012 Pitt Magazine Article about The Virtual Trillium Trail.

>>>See the Human Factors and Ergonomics Society 55th Annual Meeting Presentation for an overview of the results.

Between 2009 and 2012 the research was redeveloped for market by Virtual Field Trips, LLC.

Data Generated Virtual Ecologies for Virtual Environments in Game Engines, (2003-2008).

biophotoVirtualTrilliumTrail2004testHarringtonVirtualTrilliumTrailContorHarrington2004

An investigation into techniques required to visualize scientific terrain, plant population data sets (University of Pittsburgh, Department of Biology and long term NSF funded research), and photo-realistic 3D plants, for development of real-time interactive, immersive, environments in high-end game engines. . The design and development of interest are realistic terrain modeling and plant modeling in a statistically accurate and automatic process of nature, trees, plants and flowers.


The project has collaborated with Professor Susan Kalisz and Senior Research Associate Jessica Dunn, Department of Biological Sciences, University of Pittsburgh. The future of the research will build on the framework and can represent different kinds of front-ends for simulations. Simulations that represent different scenarios of future realities based on decision support systems and social-environmental analysis. The capability to see the results of different parameters in a high fidelity computer graphics model will be a very powerful decision support system. Unlike the more abstract visualizations, these types of visualizations will empower policy makers to understand the environmental impacts of choices. Public outreach is an important part of this research and envisioned to be a critical component of this work.

RealorVirtualVirtualTrilliumTrail2007Harrington

RealVirtualTrillium2004

VirtualTrilluimHarrington2004


The Wild Point, (2009-2012). (PDF)

WildPointHarrington2006

A virtual reality art and educational simulation of Pittsburgh, Pennsylvania, based on the historical Elias Meyer 1762 map and terrain data from Google Earth. Created to be an interactive virtual environment for both historical educational outreach and as well as a work of art. It is a visualization of pre-European America, and open to imaginative re-interpretation of the region. Valleys and floodplains are free of manmade structures, thus invites the viewer to wonder about past and future realities of the region.