Visualization is nothing new. Humans have during long time had a need to represent and externalize thoughts and ideas. Through various means, for example, pencil and paper, humans have created the opportunity to develop human thinking. Visualizations are external representations.

A tradition of visualization has its background in statistics. It involves various forms of charts that allow users to gain insights of the data that forms the basis for the visualization. Developments in technology, particularly Internet development, have meant that the area has taken something of a leap. New ways to visualize a data has its background in development of, for example, social media. One such example is the word clouds witch is a graphical representation of how often different words are used in a specific context.

Visualization is an area that for a relatively long time had the interest of researchers in the research realm of HCI but also in other research spheres – especially cognitive science. Visualization has a wide range even within HCI research. It can involve everything from visualization of statistic data, visualization to create game worlds (virtual worlds). It can be about how data from electro microscope or an x-ray machine can be visualized for the user to gain insight and create knowledge from data. The definition often used within the HCI domain is: Information Visualization – The use of computer-supported, interactive, visual representation of abstract data to amplify cognition (Card, Mackinlay, & Shneiderman, 1999).

Developments in the field have been very rapid and alongside research articles it has been important to take note of other literature books and industry reports. In the books I have read visualization is presented as an industry, branch and area of technology. The content in these books differ from the research articles I have read. This possible can be explained by that research takes time. The area of visualization has recent years attracted interest in many areas. The journalistic task is to explain something, and this technology will come in handy. Nathan Yau explains how visualization is all about “telling stories with data” (Yau, 2011, sid 1).

Two of the articles that I have read are published in other fields than HCl but have also contributed to knowledge and above all put the area in a context. Lurie & Mason describes the impact of visualization on decisionmaking (Lurie & Mason, 2007). The article is published in the research field of marketing and describes the importance of visualization to gain insights and there by be able to make wise decisions. Lurie & Mason works within the the researchfield decision-making and look up on visualization from this perspective. Lurie & Mason use the term “visual perspective” to refer to how a given visual representation changes the relationship between visual information and the decision maker (Lurie & Mason, 2007). One aspect of visual perspective is “interactivity,” or the user’s ability to change perspective, for example, by rotating or simulating movement around an a set of data. Lurie & Mason concludes that the design of visual representation as encoded information from several different aspects such as shape, color, texture, geometri. When the user, in this case decision makers, use the artifact there is a decoding. The representation works only if the coding is correct and effective and in relation to the users ability to decode.

Card, Mackinlay, & Shneiderman uses the concept of knowledge crystallixation and it referres to when one or more people gather and organize information in order to understand it or gain knowledge from it (Card et al., 1999, p 13). The result can be a report or summary. In this context it is possible to express how visualization is used to create patterns from something abstract and in this way create the ability to understanding and gain knowledge.

When reading the selected articles and literature some areas emerges. Theese are: Representation, Cognition, Data, Information, Amplified Cognition, Patters of use, Technology, Designprocess, and it is theese areas I describe in short here.

Representation

Rieber describes how visualization is a tool to solve a problem (Rieber, 1995). Visualization is a problem-solving-tool. Rieber insert visualization in the area of problem-solving research and believes that the use of visualization can be considered as a problem-solving strategy. He further describes how an external representation of a problem may be needed to find a solution.

By describing a set of simple and historical examples Rieber shows how visualization served as tools for solving problems for a long time in human history. It is not only more dramatic events in history that have a relationship with the visualizations, but often problems and decisions in common peoples everyday lives. The approach to how visualization and external representation constitute a problem-solving strategy is, according to me, well in line with user-centered design. There must be a clear thought about the problem that the visualization solve. There must be an investigation of who the user is and the benefits arising from the use of the artifact.

Rieber provides several interesting examples of the importance of being able to ”inner visualisation” The example he describes comes from the scientific comunety and how some very famous scientists such as Albert Einsten, August Kelué, Roger Shepard, all testify to how they can close their eyes and see things (for example atoms organize in a special cluster). But Rieber does not use the concept of internal representation in his text. This is a concept used in later research on visualization but I perceive that it exactly what Rieber describes with his examples. Rieber nor use the concept of external represention.This is also something that are established later in the litterature.

Liu, Nersessian, & Stasko resembles external representation with a symbol (Liu, Nersessian, & Stasko, 2008). The symbol carries meaning about something and when humans understand the symbols (as intended) a there is an effect. A picture or symbol is an external representation. A graph is a picture or symbol that represents for example a large amount of data or information. Within research area cognition the word representation mainly stands for something else namely the mental images we create in your head to understand what we are seeing or experiencing. Cognitive science solely focuses on finding internal representations and abstract rules inside the brain (Liu et al., 2008).

Liu, Nersessian, & Stasko describes how there in the HCI research community is a perception that knowledge of cognition is the responsibility of the cognitive scientists and it must be abandoned by HCI research. Even so, Liu, Nersessian, & Stasko examining a theory called ”distributed cognition” and it has its home in cognition research. It is central in this theory how humans have learned to use different tools to teach us things. The pen and paper is an example of such tools. By externalizing our thoughts, we can also take the idea further and develop new thoughts and ideas. We can also better communicate our thoughts with other people and there by collectively develop knowledge. The concept “amplify” (cognition) is launched and relate to how learning increases in power when these cognitive tools are used.

Card, Mackinlay, & Shneiderman uses the term external representation to describe how humans through use tools to relieve working memory and by this create the opportunity to push the “thoughts” further (Card et al., 1999). The written language and above all the written numbers are used as an example. It would not have been possible to develop the mathematical thinking without the ability to use symbols externalizing complex thoughts (Card et al., 1999 p. 3). The line of argument is how visualization is used for thinking and this is captured with the books subtitle ”using vision to think.”

The importance that maps has had for shipping and humans opportunity to explore the world are one given example. The development of the charts and its importance to understand large quantities of something are another example of cognitive tools. The relationship between thinking, externalization and communication is established. The researchers describes briefly how this is the case and how there is a complex interaction but Card, Mackinlay, & Shneiderman does not go into detail on the subject of cognition. Instead the goal of their analysis is to determine how complex data rendering allows for data-based visualization and how this represents a tool for thinking.

Odification, Price, Brooks, Vessey, & Street reason about how the internal representation arrangements are created and relate to external representation (Odification, Price, Brooks, Vessey, & Street, 2006). My opinion is that although this realation is very complex it is central to HCI research. My interpretation is that there is a challenge in creating a visualization that is relevant to the user. The goal for the designer is to create an external representation that can be perceived by the user and the user must make sense of it the user must gain some kind of insights from the visualization. The user must find it useful.

The basic idea is that digital technology in design of visualization creates the ability to provide different views on the same data. It is possible to present data in different ways (ways that fits the individual user and their internal representations better). In this context in the word interactivity is not mainly focusing of navigation. It’s mainly about the interaction possibilities for the user to create external representation that is in line with its internal representation. In this way the visualization becomes effective.

Cognition

There area of cognition is not described in particularly high extent in the articles I have read. There seems to be a approach with in the HCI community that “cognition as a subject shall be handle by the cognition research and I find this hard to accept. The is theories used in HCI that has its background from cognition research area and cognitive science. Cognitive fit theory is one example and it concerns the relationship and interaction between problem solving, internal representations and external representations. Liu, Nersessian, & Stasko describes information visualization as a field that not is matured and that there are a lack of theories to lean on. They see their own research as a first tentative steps regarding development of theories for visualization (Liu et al., 2008).

Cognitive fit theory provides an explanation for performance differences among users across different presentation formats such as tables, graphs, and schematic faces (Odification et al., 2006). In this manner both the internal and external representations, and the interactions among them, contribute to the mental representation for task solution that is developed to solve the problem (Odification et al., 2006). The work Odification, Price, Brooks, Vessey, & Street has done is very extensive and the article is multidimensional. Other researchers in the field of visualization are referring to this article. It is the notion of internal and external representation that most central.

Data

All data must in some way be structured in tables to be visualized (Card et al., 1999). Nathan You describe in the book Visualize This: The Flowing Data Guide to Design, Visualization and Statistics the importance of getting hold of data, structure data and format data. This is the foundation of digital visualization and Nathan Yau says that it is when the designer manages these parts as design opportunities are created (Yau, 2011).

My opinion is Nathan Yau’s book is about the possibilities new technology created the past 10 years to collect, organize, and visualize information. Not least, the book is about the technologies just to find, collect and organize data. Technology has over the years become more user-friendly and accessible. More individuals and organizations can work to visualize data.

But it is not only new technologies that are contributing to the development of the field. There are changes in society that reflects on the domain. As an example can an ongoing shift in approach in the governmental area where there is an increasingly large effort to make data available to the public in formats that can easily be used. The idea is that the data has value and that social benefits may arise if the data are easy to reach. This trend is called OpenGov (or Open Gov moment). Another area that can be singled out as contributing the development is social media such as Twitter and Facebook where large amounts of data being created and it is located in the corporate interest that their data can be used in ways that support their business models. There is a relation between the growing amount of data and the development in the area of visualizations.

Interaction

Yi, Kang, Stasko, & Jacko thinks that within HCI research there has been a focus on the representation in terms of visualization and area interactation is neglected (Yi, Kang, Stasko, & Jacko, 2007). They argue that the interaction aspect is important because if digital artifacts, visulisations being built and they are not interactive the designer doesn’t utilized the technologie. They claim there is a big difference between an analog visualization for example on paper and a static visualization (digital). Although it is understandable that the digital visuliseringen can be created ”on demand” the interaction posibilites are not used. The researchers literature review on the subject gives them a oportunety to create list of categories – different interaction possibilities:

Select: mark something as interesting
Explore: show me something else
Reconfigure: show me a different arrangement
Encode: show me a different representation
Abstract/Elaborate: show me more or less detail
Filter: show me something conditionally
Connect: show me related items (Yi et al., 2007).

The researchers go through each category carefully and also give examples. This list is the result of their research, and they argue that the list of categories can be developed. Yi, Kang, Stasko, & Jacko do not use the concept of pattern but I believe this would be termed as different patterns and the list could be very useful in a practical design work.

Amplify Cognition

Card, Mackinlay, & Shneiderman uses the term ”ampify cognition” and raising questions how and when visualization reinforce cognition (Card et al., 1999). This can be investigated by giving a group of people the same task to solve, the one given tools in the form of a visualisation while the other group did not have access to this. Then the research/designer can measure the time it takes to solve the task. Card, Mackinlay, & Shneiderman describes a few central areas where visulisations can offer efficency: Visualization offers a compact information in a small space. Visualizations can offer to zoom in on details. Trends and patterns can be quickly identified. Visalization can do so that searches are avoided and so on. What is recurrent and central is the fast accesses to information that a visualization can offer the user. Yau writes in detail about variety and recurring areas for use of for visualization. A recurring need is to describe the trends that occur over time. It is recurrent through visualization show proportions of something. Relationships between different things is grateful to visualize and spatial context (Yau, 2013).

Forms and structure of visualisations

Card, Mackinlay, & Shneiderman discuss of the notion of ”cost structure”. It relates to how the visualization is created to be as efficient as possible for the user. It’s about organizing information in a way so that the benefit for the user is as large as possible according to the users goals. The user should use as little time as possible on achieving the goals she has. Card, Mackinlay, & Shneiderman then moves on to visual structures which deals with how the visualization should look like in order to be understood and effectively used. Here comes the concept perception in.

Process

Tory & Möller concludes that the process of designing the visualization must take its starting point from the user’s needs and support the user in his task (Tory & Möller, 2004). It advocates a user-centered design process with a process consisting of analysis, design, prototype and testing then iteration of this process (otherwise, the results of this study primarily to determine that there is not much systematic research on visualization that is user-centered). Data Points: Visualization That Means Something is another book from Nathan You. Also in this book there is a focus on the importance of finding out what is to be achieved with the visualization. There is also a focus on the graphics side of data analysis using examples from art, design, business, statistics, cartography, and online media, he explores both standard-and not so standard-concepts and ideas about illustrating data (Yau, 2013).

From this book I wish to highlight how Yau expresses how visualization is a medium. The argument given is that visualization is used in so many different purposes. From statistical analysis to art, making it difficult to reason about visualization as a contiguous area. I find it interesting that it is only in this book, which was published in 2013, as visualization are described as “a medium: a way to explore, present and express meaning in data” (Yau, 2013).

It is only two years between Yau’s two books. Thematic the books are presented in a similar manner. An observation is that there are more examples in the latter book and the examples are more diverse. An interpretation I make is that the area has exploded in recent years. Yau will return in this book again and again to the concept of meaning: “Visualizations that’s means something”. The author describes a design process and a model is presented. Do note that this is only briefly done. In these books, and creates thoughts about my own upcoming project where I just will sketch out a process.

Interactive Data Visualization: New Directions for Accounting Information Systems Research was published in 2010 and has a focus on visualization and decision making in organizations (Dilla, Janvrin, & Raschke, 2010). The researchers develop a taxonomy for examining the current state of interactive data visualization research related to accounting decision making. It gives an overview on the research area and articles in the field. Dilla, Janvrin, & Raschke´s research is done with in the scientific area of information systems, but the study is published in a journal in the field of accounting. In accounting and finance, a variety of IT systems are used and from these it is possible to retrieve data. This data forms the basis for economic decisions and for the user to create insights from the data there is a need for various forms of cognitive tools. In this way, the work of creating these cognitive tools its starting point from existing methods in the field of enterprise resource planning (ERP) and Balanced Scorecard (BSC).

How is visualization for decision making used within the field of accounting? To explain this, the researcher lean primarily on two theories or perspectives. The first is ”cognitive fit” that was developed in the early 1990s by Iris Vessey (developed later, in 2006, by Vessey). The theory proposes that the correspondence between task and information presentation format leads to superior task performance for individual users.

The second theory or perspective Dilla, Janvrin, & Raschke retrieves from marketing research and it is called Judgment and decision-making (JDM) and is created by Lurie and Mason in 2007 (Dilla et al., 2010). By applying these two perspectives describe Dilla, Janvrin, & Raschke how they are developing a framework that examines linkages between task and decision maker characteristics and interactive data visualization. The framework also examines the link between interactive data visualization and decision processes and outcomes. Dilla, Janvrin, & Raschke identify conditions under which various aspects of interactivity have been shown to affect decision processes and outcomes.

As I understand it he primary conclusion is that the user’s personal characteristics play a major role regarding which kongitiva tools should be used. If tools are used where the user is able to select data, representation arrangements and type of visualization, high demands on skills and knowledge.

Most of the research discussed in this review is consistent with a cognitive fit perspective: It suggests that the effectiveness of information visualization techniques, such as interactive infor- mation and navigation tools and interfaces that allow the user to select their preferred information representation, depends on task characteristics, such as the dimensionality and complexity of data and decision maker characteristics, such as domain-specific expertise and cognitive ability (Dilla et al., 2010). As a passage describes how the user perceives well presented information and interactive information more credible. I find this very interesting because this is a wrong idea – false data or wrong data doesn’t become a truthfull or right just becaouse it is designed in a professional and attractive way.

References

Card, S. K., Mackinlay, J. D., & Shneiderman, B. (1999). Readings in information visualization : using vision to think. San Francisco: Kaufmann.

Dilla, W., Janvrin, D. J., & Raschke, R. (2010). Interactive Data Visualization : New Directions, 24(2), 1–37. doi:10.2308/jis.2010.24.2.1

Liu, Z., Nersessian, N. J., & Stasko, J. T. (2008). Distributed Cognition as a Theoretical Framework for Information Visualization, 14(6), 1173–1180.

Lurie, N. H., & Mason, C. H. (2007). Visual Representation : Implications for Decision Making, 71(January), 160–177.

Odification, M., Price, M. F., Brooks, W., Vessey, I., & Street, M. (2006). T HE R OLE OF C OGNITIVE F IT IN THE R ELATIONSHIP B ETWEEN S OFTWARE C OMPREHENSION AND, 30(1), 29–55.

Rieber, L. (1995). A historical review of visualization in human cognition. Educational Technology Research and Development, 43(1), 45–56. doi:10.1007/BF02300481

Tory, M., & Möller, T. (2004). Human factors in visualization research. IEEE Transactions on Visualization and Computer Graphics, 10(1), 72–84. doi:10.1109/TVCG.2004.1260759

Yau, N. (2011). Visualize this : the FlowingData guide to design, visualization, and statistics. Indianapolis, Ind.: Wiley Pub.

Yau, N. (2013). Data points : visualization that means something. Indianapolis: John Wiley {&} Sons.

Yi, J. S., Kang, Y. A., Stasko, J., & Jacko, J. (2007). Toward a deeper understanding of the role of interaction in information visualization. IEEE Transactions on Visualization and Computer Graphics, 13(6), 1224–31. doi:10.1109/TVCG.2007.70515