Description

Motivation

Concept maps are powerful cognitive tools. They inform and educate. They use spatial, visual, and verbal modalities to represent literal and metaphoric knowledge. They augment memory and processing of information. They enable assimilation of information as generalizations, concepts, and principles, rather than scattered cues. Recent research is pointing toward leveraging spatial, visual, and verbal thinking tools, such as concept mapping, for knowledge inquiry in education for work and life (e.g., Newcombe, 2013; Eppler 2006; NRC 2012).

Schroeder et al., (2017), in their recent meta-analysis study of concept maps, (1) concluded that “there is very little research on the specific features of concept maps that promote cognitive elaboration or reduce extraneous load”, and (2) suggested that “comparing different map designs may also be the best way to create more advanced types of concept map” (Schroeder et al., 2017, p. 449, emphasis added).

Automated node-link visualization (e.g., https://js.cytoscape.org/) can be useful to alleviate the challenge of laying out concepts (nodes) and links between concepts on a two dimensional space; a task that is widely known to be time consuming to do manually. A resulting drawing of graph algorithms depends on a number of parameters that an algorithm tends to optimize, such as topology (layout) of the drawing, edges crossing, edges bending, area, symmetries, among others (Purchase, 2002). Recent empirical research into node-link visualizations suggest that there are advantages of one choice (e.g., Huang, 2007; Dwyer et al., 2009; Lin et al., 2018). However, such studies have investigated effects of arrangements of nodes and edges where users were often tasked to perform abstract tasks, e.g., shortest path, nodes with high degree, etc., on graphs representing abstract relational information, e.g., information without much semantics.

However, in concept mapping users engage in conceptual learning, e.g., critical thinking about ideas and concepts. Qualitative investigations, for instance, using eye tracking data, suggest that users avoided dense areas where there are relatively many edges and nodes (Huang, 2005), edges crossing with acute angles confused users’ vision (Huang, 2007). It is highly probable that minimizing edges crossing, minimizing edges bends, maximizing crossing angle, maximizing symmetries will have a positive effect on users’ reading and comprehension of a concept map. However, it can be difficult to predict how the overall layout of a concept map will affect users’ when engaged in a conceptual learning activity.

Study Purpose

This study examines the effect of features of automated algorithms of concept mapping (node-link) visualization on users when engaged in reading and comprehension tasks. Empirical results of this study are important in four ways. First, it is important for researchers to understand the effect of spatial, verbal, and visual factors of concept mapping on users’ conceptual learning and/or (critical) thinking. Second, this study extends recent research into effectiveness of node-link visualizations by tackling a setting where students are engaged in conceptual learning activity, rather than abstract tasks and/or abstract relational information. Third, this study might also be important for teachers to understand how unique design features of a concept map might enhance students’ reading and comprehension. And finally, the study might also be important for education specialists and designers to inform the design of concept maps to best support users’ readability of conceptual information.

Method

This study makes use of existing software libraries for creating concept map visualisations (see the figure) and follows a mixed-research design for hypothesis testing.

- Hypothesis

Depending on the direction that the student would like to take, examples of hypothesis to examine include, but not limited to:

H1: Will different types layouts (e.g., radial, circular, grid..) have an effect on users’ reading and comprehension of a concept map

H2: Will different types of signalings (e.g., colors, landmarks, images,...) will have an effect on users’ reading and comprehension of a concept map

H3: Will different sizes of a concept map (e.g., small, medium, large,...) will have an effect on users’ reading and comprehension of a concept map

- Content and Tasks Design

We co-design the tasks (content + concept maps) with teachers or we engage with teachers to review and provide us with feedback about the tasks.

Tasks will be designed to examine the hypothesis.

- A Test Bed

Designing a test bed (e.g., https://js.cytoscape.org/) and perform a usability study using eye tracking.

Participants will be tasked to perform reading, comprehension, and/or concept mapping tasks.

- Data Collection

We will collect users’ task time completion, accuracy, preference, eye tracking, data to test the hypothesis.

- Data Analysis

Data will be statistically and qualitatively analyzed to validate the hypothesis.

Research Team

- You will be joining an interdisciplinary research team, collaborating closely with a postdoc member of the team.

- We hope to have another student to collaborate with us on a related MSc topic (see here) to compare and contrast the results of both projects.

- You will be encouraged and guided to develop a strong research stance on concept mapping design and use, to further motivate the hypothesis for the study.

- You will be supported and guided throughout this project to pilot-test all the aspects of the study, design of the content, tasks, recruit participants, conduct the study, collect and analyze data, and report and hopefully publish the findings in a journal or a conference.

Objectives and Methodology

Examine the effect of features of automated concept mapping (node-link) visualization algorithms on users’ conceptual learning: reading and comprehension of concept mapping. For the study, the student will be encouraged to use learning materials, for instance, from the core module (The Core of GI Science) from the Living Textbook; to create different concept maps’ designs and learning tasks to evaluate with users in controlled experiments (see “method” below). It is important to take into account the users’ perception of space and spatial relationships in the concept map, such as connectivity, proximity, clustering, etc. in order to optimize the visualisation in the concept map.

Further reading

References:

National Research Council. Education for life and work: Developing transferable knowledge and skills in the 21st century. National Academies Press, 2012. https://doi.org/10.17226/13398

Newcombe, Nora S. "Seeing Relationships: Using Spatial Thinking to Teach Science, Mathematics, and Social Studies." American Educator 37.1 (2013): 26.

Eppler, Martin J. "A comparison between concept maps, mind maps, conceptual diagrams, and visual metaphors as complementary tools for knowledge construction and sharing." Information visualization 5.3 (2006): 202-210.

Schroeder, N. L., Nesbit, J. C., Anguiano, C. J., & Adesope, O. O. (2018). Studying and Constructing Concept Maps: A Meta-Analysis. Educational Psychology Review, 30(2), 431–455. https://doi.org/10.1007/s10648-017-9403-9

Purchase, H. C., Carrington, D., & Allder, J.-A. (2002). Empirical evaluation of aesthetics-based graph layout. Empirical Software Engineering, 7(3), 233–255.

Vismara, L., Di Battista, G., Garg, A., Liotta, G., Tamassia, R., & Vargiu, F. (2000). Experimental studies on graph drawing algorithms. Software: Practice and Experience, 30(11), 1235–1284.

Huang, W., & Eades, P. (2005). How people read graphs. Proceedings of the 2005 Asia-Pacific Symposium on Information Visualisation-Volume 45, 51–58.

Lin, C.-C., Huang, W., Liu, W.-Y., & Chen, W.-L. (2018). Evaluating aesthetics for user-sketched layouts of symmetric graphs. Journal of Visual Languages & Computing, 48, 123–133. https://doi.org/10.1016/j.jvlc.2018.08.004

Dwyer, T., Bongshin Lee, Fisher, D., Quinn, K. I., Isenberg, P., Robertson, G., & North, C. (2009). A Comparison of User-Generated and Automatic Graph Layouts. IEEE Transactions on Visualization and Computer Graphics, 15(6), 961–968. https://doi.org/10.1109/TVCG.2009.109

Ware, C., Purchase, H., Colpoys, L., & McGill, M. (2002). Cognitive Measurements of Graph Aesthetics. Information Visualization, 1(2), 103–110. https://doi.org/10.1057/palgrave.ivs.9500013

Huang, W., Hong, S.-H., & Eades, P. (2007). Effects of Sociogram Drawing Conventions and Edge Crossings in Social Network Visualization. Journal of Graph Algorithms and Applications, 11(2), 397–429. https://doi.org/10.7155/jgaa.00152