Visualization in Learning: Perception, Aesthetics and ...

1

Visualization in Learning: Perception, Aesthetics and Pragmatism Veslava Osinska Institute of Information Science and Book Studies Nicolaus Copernicus University, Torun, Poland Grzegorz Osinski Institute of Computer Science College of Social and Media Culture, Torun, Poland Anna Beata Kwiatkowska Institute of Mathematics and Computer Science Nicolaus Copernicus University, Torun, Poland

ABSTRACT Visualization is currently used as a data analysis tool and considered a way of communicating knowledge and ideas in many areas of life such as science, education, medicine, marketing, and advertisement. Authors try to show that information visualization techniques are being more and more widely applied in the education process. Visualization mechanisms are designed taking into account analytical and contentrelated potential, timeliness, online availability, and aesthetics. The classical (tabular) forms remain dominant in information presentation. They can be observed in the structure of most websites and elearning courses. Alternative solutions are non-linear layouts e.g., network or fractal. Visualization maps with specifically designed architecture begin to play an important role in education and science development. Authors emphasize that implementation of such tools should be supported and developed in e-learning platforms. To create a knowledge map we need to use advanced data analysis and layout algorithms. Mapping of information requires interdisciplinary collaboration between researchers in different fields who can perceive and apply contemporary trends in visualization including natural shape perception, 3D representation problems, as well as the aspects of neuroaesthetics. Keywords: information visualization, visual perception, cognitive approach, visualization in education, fractal graphics, fractal structures.

INTRODUCTION Visualization methods are currently used for scientific and scholarly presentations and considered the communication tools in interactive web applications. On the other hand, we must take into consideration an increasing role of mobile technology in communication processes. Visualization methods in education are still underestimated. In most cases numerical data are presented in a tabular form or by twodimensional graphs and charts. An extended gap has appeared, between classic forms of information presentation and the users (students) who utilize new technologies with a particular emphasis on mobile devices. It could be observed during e-learning processes where an emotional component provided by the direct communication is lost. Introducing special graphic modules strictly related to new achievements in cognitive science should fill this gap.

2

Usability of visualization mechanisms depends on several factors, namely their analytical and contentrelated potential, timeliness, online availability and, of course aesthetics. Many information visualization (Infovis) projects available on the Web meet these criteria. They provide their users with instant feedback and data manipulation options and offer social collaboration in visual analysis. Virtual exhibition Places & Spacesi is an interdisciplinary portal for researchers concerned in scientific domain mapping and human activity across global history. The visual explorer IBM ManyEyes ii allows the users to visualize their own raw data and finally share results and interpretations. In this chapter, the authors try to show that data visualization techniques are being more and more widely applied in the education process. Being not only simple forms of visualization, but also colorful and often interactive maps, they become a perfect teaching tool in the education process. Visualization is not just a methodology that originates from computer graphics and data analysis and is applied in many areas of life (science, education, medicine, marketing, etc.), but it is also an effective and popular way of communicating knowledge and man’s ideas. Thus, visualization is becoming more and more important for today’s users – students – readers – consumers. This chapter contains complex interdisciplinary material and attempts to construct a general framework of the role of visualization in learning. This diversity of content requires a special (non-linear) form of representation. Instead of a classical table of content, the conceptual map with graphical explanations is included in Figure 1. The map shows the issues discussed in the chapter and the history of their origin – first that in the authors’ brain and then on paper. The sequence, mutual connections, and similarity between specific issues are presented by arrows or close location. Based on visualization study, the various aspects of perception of information structure are discussed, as well as science maps and visual elements used in education. Special attention is given to the creation of new qualities of knowledge structures, and dynamic processes in the space of the mind (Duch 1998). Figure 1. Veslava Osinska. Content map (instead of a table of content of current chapter). (© 2014, V. Osinska. Used with permission). [Figure 1 about here]

3

Image perception is here discussed from the user’s point of view, including both ergonomics of placed information resources and the aesthetic structure of presented images. In recent years many works on that subject have appeared and a rich collection of results from scientific experiments is already available. Fractal structures are exceptionally intuitive in perception and reception because they originate from (or resemble) nature. This explains why fractal-like visualizations are perceived better. Visual communication messages should be constructed by following such patterns. This neuroscience-based issue has been solved in nature by fractal structures that are easy to compute in iterative way and reflecting the structural complexity in the form of aesthetic communication. Discussion focuses on visualization methods as a tool for building a bridge between natural perception and educational materials construction. This matter is currently used in scholar presentations and considered to be communication tools in interactive web applications.

BACKGROUND Visualization as a Map The term ‘map’ is usually associated with cartography and those are the first maps that we come to know. They describe the world at a two-dimensional level using a specific language. They use a plane colorful image to show a great amount of information about the world that we cannot see directly, but which we can imagine. In the space of our mind we can create an image of the world presented on a map. Children make their first journeys as ‘armchair travelers.’ This is how they get to know names of places, identify types of landscapes and distances. Technical geographic maps provide information on climate, economy,

4

social and economic conditions, and many other details. The world that has been confined to a piece of paper is an indispensable element in the process of school education all over the world. However, maps abandoned long time ago their original geography-related environment. First, they naturally appeared in history related sciences where, in the same way as in geography, they described the longer existing worlds, and showed no longer present towns and societies. Nevertheless, in the learning process during classes or individual learning these no longer existing worlds come to life in students’ minds, so they can identify themselves with places and events from the past. Every teacher knows that analysis of a complex problem should start from working with a map, which provides an insight into the whole problem. Maps enable a detailed scrutiny of each element in reference to various factors, both these included in the map and these that students have learnt before. The effectiveness of teaching and learning with the use of good maps is unquestionable, and every teacher and learner realizes it. How can classic mapping techniques be used now in the 21st century, and serve for teaching and learning not only geography or history? In the early 1980s, atlases related to many kinds of science started to appear, besides geographical and historic ones. They related to philosophy, mathematics, physics, chemistry, and many other domains. In a classic way, maps and graphs attempted to show the entire systems and methodologies of a given science field. However, they were encyclopedic in nature; it was difficult to find a science map similar to the geographical one that would be not only interesting to students but also enable them to make individual multi-variant analyses of the presented content.

Problems related to Computer Visualization The development of computing methods has enabled to facilitate and improve that process. Visualizations made on paper include such limitations as a lack of interaction, so the users cannot manipulate the data nor play with their different configurations. Computers have introduced a lot of possibilities and challenges into communication with users. The first interactive geographical atlases had already appeared before the GPS technology became popular. They allowed dynamical viewing of maps on one screen. Limitations resulting from a two-dimensional nature of paper disappeared, but other came into view. Excess of information – redundancy – became a problem. Dynamical maps contained so much information that the user could feel lost. A great amount and diversity of information hindered a quick analysis of a specific problem. The information factor outweighed the analytical one. First multimedia computer atlases basically became huge lexicons of data, except that the text there was more often substituted by image, film, or sound. The introduction of the GPS has actually only exacerbated the problem. Maps include information from the real world, which is however ephemeral and temporary in nature, such as information on restaurants, gasoline stations, or tourist attractions. This does not contribute much to the analytical and creative problem solving. In spite of the application of computer technologies, the potential analytical capability of maps has not increased, only their informative nature expanded. The situation is similar in other sciences where computer atlases – map collections – have expanded their informative content, but have not created a new analytical quality.

Educational Role of Maps The application of maps in education has an entirely different nature than just a medium for a large collection of information. Everyone knows that reading maps is something one has to learn. Current process of tabloidization of information causes that the analytical elements and the semantic image descriptions are becoming insignificant. The whole information is to be communicated directly, hence a great popularity of pictograms, even for expressing feelings. Emoticons used by users in text messaging have replaced reflection on one’s own emotional state. The Technical Report of the OECD’s Programme for International Student Assessment (PISA 2009) shows that in many developed societies of Europe and the USA over 30% of society finds it difficult to understand weather maps presented in everyday TV

5

news. One can actually notice a growing simplification of communication forms. Five years ago, many TV channels would present weather maps with isobars and isotherms that showed a direction and type of weather change. Nowadays, more and more weather maps include only primitive icons of sun, clouds, or rain. It does not solve the problem but only makes it worse. Analysis of a well prepared weather map not only enables us to understand whether it is necessary to take an umbrella to work the next day, but also let us know should we expect weather changes during the following days, whether we will be affected by continental or maritime air masses, and many other details.

Cognitive Aspects From the cognitive point of view, the problem is that perception system is excessively loaded with a great number of visual data, with a simultaneous disconnection of modules responsible for content-related analysis. A similar problem has been noticed in the development of e-learning technologies. Too much information given at the same time and often too easy access to that information result in users being passive in their individual attempts at information analysis and remaining ‘viewers’ but not ‘readers.’ It is a typical behavior for the generation that grew up in a TV culture without a possibility to interact. The reception of a map should be comprehensive, treated as a global resonance between a preconception in the user’s brain and the presented graphic content (Duch & Grudzinski 2001; Duch, 2007a). Simply speaking, map should make us think. In order to make it possible, the user should be taught how to read maps. That is undoubtedly an issue related to the education field and it requires an interdisciplinary approach of many specialists. Education system should provide the separate education paths, which would be directed towards a deep analysis of the presented content, not just towards reading simple messages. Therefore, special attention should be given to the application of preconception systems that already exist in users (Duch, 2007b). Preconceptions are our intuitive and stereotypical ideas about processes and objects that we have never reflected upon, but which have appeared spontaneously during the course of our life. Sometimes we do not realize that they exist, but a specific problem that we face activates those processes in our brain in the first place. When we walk around any art gallery and stop in front of a selected painting, in a natural way we think about what that painting reminds us, and then we start to analyze the message it carries. A painting always carries an informative and emotional load. The primary mistake of many educational methods is that they omit or diminish the meaning of the psychological factor. It is, however, the emotional load that activates these preconceptual meanings, which are crucial to the user. Childhood memories or personal experience are more significant than a semantic description that often accompanies a presented image.

Aesthetics of Visualization The aesthetics of the graphic communication is becoming a highly significant element that must be adjusted to an individual user. It is especially important during map reading. Different colors and a selection of textures can be very attractive to one user, but off-putting to another. It is difficult to select a universal graphic design, but map presentation technology enables a wide application of various techniques as well as their combination. If that specific syncretism is used within a proper scope, it will probably help to find the golden mean in selection of an adequate graphic design of a presented map. It is vivid in the versatility of applied techniques, e.g. the Places & Spaces exhibition. However, there is no any easy recipe for a proper selection of a graphic design. Application of principles resulting from the Gestalt psychology is of course necessary but definitely insufficient. It is difficult to assume that a process of the whole image perception is based just on a simple analysis of shapes, lines, location, and configuration of specific shapes. The role of the limbic system in the image perception process is often underestimated, and this influences the effectiveness of didactic communication. What still remains a great mystery is a huge emotional load that is stored in human brain since the early childhood, and which has a decisive influence on the perception of the world around. We just ‘do not like’ some shapes and

6

colors. Different shapes can activate feelings of rejection or fear in our brain. On the other hand, positive associations are adopted easily; they activate cognitive processes that can activate analytical processes. If the world around us does not have simple Cartesian shapes, why are students would be expected to show interest in quadrilaterals, tables, lines, and rows of figures, which are often presented in course books and educational presentations? We should turn towards non-linear methods that show the whole shapes and textures, which resemble real shapes. They present data in the form of multi-dimensional graphs that should be deeply analyzed. Such thorough analysis always requires a specific amount of time. Nowadays however, immediate solutions are important, and students expect a quick solution to a problem. Can these two seemingly contradictory goals be combined so that a quick creative analysis of a considerable database is possible? It seems that application of visual methods can facilitate the achievement of the assumed goal.

VISUAL PERCEPTION Vision of Shapes or Vision of Structures? An analysis of visualization is impossible without taking into account the mechanisms of perception occurring in neural correlates of the human brain (Duch & Diercksen, 1995). Very often, in their works on visualization authors focus on a model of perception directly resulting from the principles of the Gestalt theory. The assumptions of that theory, especially ‘laws of grouping’ are presented as fundamental principles of visual perception of complex objects. The success of the application of the Gestalt Law is unquestionable, both in some applications in computer science, especially in designing graphic interfaces, and computer vision systems. However, in the process of education it is more important to activate higher cognitive processes by involving visual perception mechanisms (Figure 2). Figure 2. Veslava Osinska, Visual cognitive processes scheme. (© 2014, V.Osinska, Used with permission). [Figure 2 about here]

It is crucial to understand how not only new impressions, but also new ideas appear in the brain. These processes are undoubtedly dynamical, and a common term ‘shining’ shows that in the final stage of creating a new idea they are remarkably quick. Neural-dynamical concepts, which result from psychological processes examined with neural imaging show a close relation between visual perception

7

and the activity of neural correlates in different parts of the brain – not only in the cortex area responsible for visual activity. These processes are characterized by very quick dynamics, and their precise examination would probably require application of experimental methods combining genetic engineering, nanotechnology, and fluorescent optics (Alivasatos, 2002). The study of dynamic conditions of neural correlates is a fast developing discipline. These sets of neurons in the human brain take part in various processes, but time-related evolution makes some of their groups create permanent schemes of links, which is manifested as repeated sequences of arousal; they are called attractors (Cossart, Aronov, & Yuste, 2003). Attractors can become basic entities describing the behavior of human brain when thinking, remembering, or decision-making. However, they change so quickly that the currently available experimental methods are not able to investigate them. Human beings can experience visual impressions as an immediate resonance between a visual pattern and the attractor structure existing in the brain. In the case of complex graphics, which any color map definitely is, it is not just the shape but also the structure of a perceived object that may be important during visual analysis. There are many open source applications, which allow users to study an irregular structure of a big dataset. For example, Gephiiii is free software package for networks and complex systems providing visualization and exploration in an impressive and instant way. Like in case of Photoshop but with reference to data, the users interact with representations; they manipulate structures, shapes, and colors to reveal hidden properties. In further chapters authors present its features in more detail. The Gestalt theory, which deals with the shape perception, will not be useful in solving that problem. As early as in the 1950s scientists advocated an approach that was different from the classic theory of the description of shape perception (Allport, 1955). However, if we look at our personal experience, we can recall situations when we spent a lot of time trying to discern a small detail in the analyzed image, but we were not able to see anything; and the other way round, we could see a small difference in the image that nobody else was able to see. It was definitely the ‘fault’ of our internal attractors; if it were possible to identify and train them in the right way, our perception abilities would surly improve incommensurably. Unfortunately, we are not able to do that yet; it is necessary to focus on methods that would help to obtain this ideal state. Applying „blindly” these methods would force the brain to seek the most efficient solution. How does visualization influence the user’s understanding? To answer this question we should take into account mechanisms of perception occurring in neural correlates in the human brain. Both an eye and a visual cortex are powerful, parallel, wide-bandwidth processors directly coupled with cognitive centers of information processing (Necka, Orzechowski & Symura, 2006). This indicates that vision and thinking are closely connected during exploration of the world, and these two aspects are the reference points in the visualization research.

Imitation – Creation as a Cognitive Processes Natural human ability to „see” causes that the brain receives projection of an object from the real life, which initially serves as the imitation of reality. Next, in a process that uses various modules, the brain compares that imitation with the available familiar objects. After that, in the space of the mind there appears a new entity that is recognized and manifested as the perceived object. Any theory of visual perception will be incomplete if it does not take into account the problems resulting from linguistic representation that semantically describes the perceived image. To understand this, try to search in the Internet browser not a text but a picture that you can well remember as a visual object. Let’s assume that you remember a photo of you and your friends that was taken while sailing on a lake. You remember that you were standing next to each other against a yacht named ‘Vision Queen.’ The water in the lake was light blue, and you could see some parts of the marina in the background. You can remember that photo very well, but you are not sure about the names of your friends, and you cannot remember if it was five or six years ago. How will you start looking for that

8

photo? If you enter into the browser the words describing the view of the photo – yacht, lake, blue water – these words will unfortunately describe millions of similar photos taken all over the world. You may of course, limit the scope of searching to a given geographical place by entering the name of the lake, if you remember it. However, it is usually unsuccessful. A semantic description of a photo does not always enable finding the correct image. Nevertheless, the brain can do it very well. It first compares familiar images, and then ‘attaches’ (although not always) semantic descriptions to them. Don’t you have in your album any photo of people and events that you actually remember, but you are not able to name the persons or objects? This shows that in the brain there is usually a stronger visual representation than a consequent semantic description that appears in the stage of image description and analysis. Although we think that these processes take place simultaneously and have common features, activating both the specific parts of the brain and cognitive activity related to the perceived objects, they are indeed separate processes that should be investigated based on various models of behavior. The process of perception, although not yet entirely investigated, can be studied by means of computer simulations. They are based on the discoveries in the field of human brain construction and functioning. Application of computer technologies enables the study of natural process of perception and recognition of real complex objects (Barres & Lee, 2014). Based on the simulation results and studies from experimental psychology, it is possible to present the whole process schematically (comprehension) in the form of a simplified scheme (Figure 3). In the presented model the visual cortex modules responsible for shape recognition are activated first. Representation of the observed objects is formed in Visual Memory (VM) where interconnections and their mutual arrangement exist. These objects relate to the memorized shapes stored in Long Term Memory (LTM). They finally create a memory structure that reveals the semantic content. Artefacts appearing in that place can resemble an observed shape, but they do not reflect its semantics. This is the base of optical illusions. Figure 3. Veslava Osinska, A model that shows processes occurring during map reading. Adapted from Barres and Lee’s recent work (2014). (© 2014, V.Osińska, Used with permission). [Figure 3 about here]

9

While reading a visualization map, cognitive perception structures are activated by means of analytical modules in a natural way, in contrast to the forced use of verbal narrative alignment. By analyzing the scheme presented above and building on the authors’ own study (Osinska, Dreszer, Osinski, & Gawarkiewicz, 2013), it is possible to draw conclusions about the process of reading and analyzing science maps. The most striking is the searching for familiar problems, forms, or contents. When the scientists view a map, they look for their own disciplines, but students focus their attention first on vivid and colorful presentations and only then on the content. Such behavior is congruent with the perception model. Scientists have a vast knowledge acquired for years, so they instinctively understand presentations that match the already existing knowledge structures. On the other hand, students first react according to the ‘first impression’ rule, and then try to understand and analyze a selected map. This observation is an important indicator for designers of maps and school illustrations. Graphics should be constructed in a way that the first impression effect is used, and directs the observer’s attention to the further path of deep analysis. Web designers and information architecture specialists know and successfully use these rules in their weighty projects. But in the era of the network big data it is not enough to attain full functionality of web applications. Knowledge about current visualization techniques and their integration with content architecture contributes to the social success of web services.

VISUALIZATION METHODS IN INFORMATION ARCHITECTURE The major problem in practical visualization is data structure’s presentation. The natural way to display a set of numbers is to place them in rows and columns, i.e., make a tabular arrangement. Tables, as the basic form of quantitative dataset representation were known in ancient times. According to some sources (Few, 2009), first tables were invented in Egypt to record astronomical observations. Greeks used first multiplication tables. Today tabular presentations remain fundamental in information architecture, particularly in Web services. Most of the contemporary website layouts are organized on grids. Such pattern is characterized by high functionality and legibility because it enables the use of modular units, which can be developed by editors, specialists and members of other teams. Website templates, which are usually classified by themes, graphic styles, or color schemes, consist of these constructing blocks. There are also conventions of organization hierarchies regarding the content. Users should feel comfortable with navigation through the web service structure, as well as exploration and searching its resources. It means that a design should be consistent and predictable in terms of users’ needs (Lynch & Horton, 2008). A table as a standard in the Web architecture meets these criteria. Tables expose non-linear cognitive properties due to vertical and horizontal parallel arrangements of data. Users can comprehend the change and the relationship between data only if they perceive sequential or selected items from the dataset in one or more groups. How does this part of the data relate to that part? How are they similar, and how are they different? What meanings do the data carry if they are grouped or taken apart? Ralph Lengler and Martin J. Eppler (2007). Presented visual complexity of visualization methods in the form of a structured table. The project is called: A Periodic Table of Visualization Methods (Figure 4). The inventors classified properties of Infovis techniques according to a rational key. They coded them both by color and position associated with a chemical element in a periodic table of elements. The authors expressed an “effort of defining and compiling existing visualization methods in order to develop a systematic overview based on the logic, look, and use of the periodic table of elements.” Thus, it is possible to say that tables allow perception of information in a non-linear way. Designers try to extend the advantages of this simple and intuitive organizational form. Rows and columns can be interchanged in pivot tables in Excel, which additionally provides a dynamical summarization of the grouped data – the name of the table comes from pivoting mechanism. It is worth noting the differences between data visualization and information visualization. Data visualization is focused on measurable data, such as human body medical examination results or geographical information system data. Information visualization, on the other hand, is concentrated on the

10

abstract data such as text or hierarchical structures. The actual information representation engages knowledge of natural ability of a human being to quickly recognize images. However, not all information can be reduced to its direct interpretation in the physical world. Figure 4. Veslava Osinska, An example of tabular organization of information. The set of known visualization methods takes form of a periodic table. Each component (cell) represents a method/technique of visualization. A detail adapted from an interactive Web application “Periodic Table of Visualiation Methods” designed by Lengler & Eppler, (2007)iv. (© 2014, V.Osińska. Used with permission). [Figure 4 about here]

Infovis is a scientific discipline that seeks new graphic metaphors to present information, which lacks a natural and obvious representation. It uses the achievements of related disciplines, such as scientific visualization, data mining, human-computer interaction, vision, visual perception, and computer graphics. The idea of this discipline is well reflected in the following quotation: “The eye seeks to compare similar things, to examine them from several angles, to shift perspective in order to view how the parts of a whole fit together” (Luther, Kelly & Beagle, 2005). Such a visual analysis is focused on the process of understanding and discovering the meaning of data. Visualization techniques are used as one of the most effective forms of data mining; they can usually find more correlations than typical statistical methods. Some scientists consider information visualization in the context of knowledge management as a stimulus to its understanding. Then, a definition in relation to that concept is: “Infovis – it is the process of knowledge internalization by the perception of information” (Dürsteler, 2007). As a result, understanding can be interpreted as a continuum, which extends from primary data to wisdom, passing through information and knowledge combined in the process of visualization. Data are simple facts; when separated from the context they lose their potential to create knowledge representation in the user’s mind. However, when formed in a proper structure adjusted to basic principles of human perception, they become more than just facts – they acquire a proper representational context, and in combination with an aesthetic feeling they gain a new cognitive quality. Data become information provided that they carry the information that we are able to understand and perceive as valuable. Together with the development of computer technologies, engineers and scientists were intensively looking for new effective methods of data structure visualization. Hierarchical trees were not represented as branches, but maps; one-dimensional typology was expanded to two dimensions. Ben Shneiderman (2009) introduced in the early 1990s a tree-map concept, an approach to the information visualization of hierarchical structures. Interactive data visualization software was named TreeMap (Figure 5).

11

Figure 5. Veslava Osinska, Representation of hierarchy as a traditional tree design and contemporary treemap (left). Screenshot of Treemap’s application for file structure management (right). (© 2014, V. Osinska. Used with permission). [Figure 5 about here]

The tree map solution is based on nesting rectangles where smaller rectangles’ areas are proportionate to the capacity of folder resources. Another idea to move the catalogue tree structure into the twodimensional space is a hierarchical scheme sketched by means of concentric circles, in the middle of which there is primary information content while other concentric circles represent related information fields. Such characteristics as general information content and type of data record are identified by segment angle and color, respectively. Another way of visualization includes creating special maps, which show not only hierarchies but most importantly the structure. An individual structure can show a natural dispersion of elements of a presented set. An example of such a map related to the visualization of different branches of computer science is presented in Figure 6. Figure 6. Veslava Osińska, Computer Science literature visualization according to main 11 thematic categories represented by colors. The map represents classified articles from ACM digital Library. (© 2014, V. Osinska, Used with permission). [Figure 6 about here]

12

Network science can also be applied – objects are presented as nodes and relations between them as edges. The most popular tool for network drawing and exploration is Gephi – open-source software for interactive visualization of large collections of data. Intuitive interface provides different layouts based on force directed, multi level algorithms, statistics, and metrics of framework, as well as manual manipulation of graph and data. As a result, multicolor graphs appear with complex structures, which constitute an example of large-scale data visualization that is easy to read (Figure 7). The analogy of such structures to fractal shapes in the context of visual perception is discussed in sub-chapter below. Graphic description of such difficult issues as the development of concepts and paradigms in the development of science is an example of the network approach to information presentation. Science maps, as a useful tool not only in education, but also in the work of professionals, allow investigating mutual thematic relations and making conclusions on the complementarity of the interdisciplinary research. In the era of unification of scientific research that we now live in, science maps make a priceless tool. The 19th century division of sciences is sinking into oblivion. Figure 7. Veslava Osińska, Social network graph, generated in Gephi. (© 2014, V. Osinska, Used with permission). [Figure 7 about here]

13

Internet technologies have not only improved the access to research findings in various scientific circles, but most importantly they have allowed those circles to become widely active in different science fields. Similarly, users such as students are less interested to which discipline a problem they analyze belongs, than how to solve that problem. Science map mining (Figure 8) enables an analysis of trends in future directions of research development in any region of the world, and – what is significant – their comparison. Patterns appearing on the maps of disciplines and scientific specializations prove a growing interdisciplinarity of science. It should be reminded that according to Thomas Kuhn (1996), the creator of the concept of paradigm shift in science, typical scientists are not “objective and independent thinkers, but conservatives” because they apply knowledge that the theories they have been taught dictate. On the contrary, users treat information in a practical way. It is up to the map designers whether their maps will be presented to users in an aesthetically attractive way or whether users will be “closed” in the dry structures of tabular data in a classic way. Similar maps encompassing different disciplines and using different graphic representations can be found in the Places & Spaces project. On the other hand, while seeking new solutions, science researchers usually analyze the contents of scientific literature. They should not limit themselves to the traditionally defined scope of a given discipline or tested and accepted methodologies. The study of sources coming from thematically remote sciences may facilitate both finding surprising answers and broadening horizons. This, in turn, may definitely have a direct influence on potential discoveries of new principles and theories.

VISUALIZATION IN LEARNING

14

Visualization of knowledge The history of human activity shows that visual representations support communication of ideas, concepts and information. To explore data and to “examine them from several angles of view to shift perspective in order to view how the parts of a whole fit together” (Luther et al., 2005), not only the eye but also the cerebral cortex are engaged. Visual representations facilitate perception of meanings of the data and relationship between their parts. If application interface is interactive, it additionally amplifies human cognition due to possible actions such as data sorting and filtering. Generally speaking, visualization is graphic information with some abstraction level produced artificially – not naturally such as in a physical landscape. Abstractions can be easily recognized. Visual forms substitute the textual content. They have intuitive attributes such as color, shape, size, position, texture and luminance. The initiator of graphics semiology, a French cartographer Jacques Bertin, defined six attributes as main visual variables, which are the basic blocks of human vision. Finally, visualization can map the data into the space of visual attributes representing them in meaningful way. These visual characteristics express the importance and dependences between data. End users apply them for the analysis of data from different perspectives. One could say the purpose of visualization is to help users acquire knowledge. That property has been known for a long time and applied in education as a supplementary learning tool. Visualization is indispensable for a teacher who wants to show something that cannot be observed in the surrounding world. So, computer simulations based on 3D mapping and rendering algorithms help the physics teachers demonstrate micro objects, such as atomic structure, or interaction of molecules, as well as macro objects, for example galaxy birth or black holes origin. Historical simulations introduce a prehistoric era and also show how borders of different countries changed in time after wars or under critical geopolitical conditions. Visualization can be used to track past events such as dinosaurs or to look at things that are difficult to reach like a human skeleton. Since the National Science Foundation supported the initiative to encourage computer graphics to visualize the inside of human body, 3D imaging has become one of the dominant directions in visualization (McCormick, DeFanti, & Brown, 1987). Thus, to prove the usefulness of visualization to academic audience, one should mention such scientific domains as: History, Anthropology, Geography, Anatomy, Medicine, Physics, Biology and Astronomy. Mapping techniques such as mind maps are useful for representation of ideas and knowledge but also for quantitative data visualization. When numerical information is presented visually, it gains a form, which allows users to have an insight into the data and catch differences, trends, and exceptions. The traditional line or bar charts can provide a combined picture of relationships between data. It would be impossible to exhibit these characteristics from the same data presented textually. Gapminderv is user-friendly software for big statistical data visualization, which supports understanding the world developments in socio-economical contexts. It is a useful tool for teachers, because it provides a fact-based worldview by means of clear bubble charts animated over time, and offers free collections of statistical data. Figure 8 illustrates the screenshot of a Gapmanider interface. Students can interactively compare selected countries by economy, education, policies, and societies, and examine how they have changed during the last 200 years. For example, it is possible to conclude that (in 2007) math education in Japan is very effective in contrary to that of the USA, despite that they have similar income per person. To explain functions of visualization, researchers often use the term insight. Seeing visualization in connection with insight is the major reason for associating it with cognitive processes. Unfortunately, there is no clear-cut translation of this word into some other languages; scientists map its meaning into a set of terms related to intuition, deep observation, careful look, or instant analysis or just perception.

15

Figure 8. Veslava Osinska, Math education achievement vs income per person for selected countries in 2007. Screenshot from GapMinder. (© 2014, V. Osinska, Used with permission) [Figure 8 about here]

Maps can serve for making a complementary analysis of the rate of changes in non-numerical information. Traditional linear data representations – charts – are not sufficient tools in this case. Maps give multi-perspective analytical possibilities and larger interpretative flexibility (Rafols, Porter, Leydesdorff, 2010). Science, as a highly dynamical, varied and unpredictable entity, is often employed for bibliometric mapping. Science maps are used to describe how specific disciplines or research fields are structured. There are some directions of the mapping study. One is focused on the representation of collaboration ties within a scientific community. Another one attempts to show the structure of disciplines. There is also research oriented at science dynamics and detection of future trends. The development of a scientific domain including its researchers is analyzed conceptually, intellectually, and socially. A lot of useful visual resources regarding this topic are available on a collection of scientific maps Places & Spaces, Mapping Science (scimaps.org). One can find there conceptual, domain-related, and classic cartographical maps of science. It is not a trivial task to group knowledge visualization maps that are varied in relation to the subject, methodology, data, and authors’ majors. The exhibition team guided by Katy Börner (2012) categorized those maps according to specifications like ‘Visual Interfaces to Digital Libraries,’ ‘Reference Systems’ or end users, for example: scholars, economic decision makers, and children.

16

A characteristic example of a science map explored on the Places & Spaces service is the so called UCSD (University of California in San Diego) map made by Richard Klavans and Kevin W. Boyack (2006, 2007) is demonstrated in Figure 9. This is the most frequently cited and demonstrated map of worldwide scientific knowledge, probably due to the completeness of the map in terms of bibliographic data. The map is used globally and universally to visualize the current state of science. Its additional virtue is that the authors have modernized it using an extended dataset and developed an interactive version so that users can play with different graphic configurations on a spherevi. This spatial graph shows the global science divided into 13 color-coded disciplines. They are (from the left): Computer Science, Mathematics and Physics, Engineering, Chemistry, Earth Sciences, Biology, Biotechnology, Medicine, Infectious Diseases, Health Science, Brain Science, Social Sciences, and Humanities. The highest density indicates a rapid development of Medical Science and Brain Sciences. Computer Science and Engineering are most commonly used in medicine. Social Sciences and Humanities are linked to Computer Science. An increasing use of information and computing technology (ICT) in Humanities is very vivid. Epidemiology as a separate category of Medical Science reveals the greatest isolation in relation to other disciplines (Börner, Klavans, Patek, Zoss, Biberstine, Light, Lariviere, & Boyack, 2012). Figure 9. Kevin Boyack. Visualizations of the UCSD Map of science: 2D Mercator projection (left) with three 3D spherical insets (top), 1D circular map (right). (© 2012, K. Börner et al. Used with permission). [Figure 9 about here]

When considering numerous examples of multivariate data mapping in the Places & Spaces collection, the authors relied on observations, assuming that the maps of knowledge and the maps of science could extend and make e-learning methodology more effective. Science maps, as an example of multidimensional presentation on a simple paper surface, constitute an interesting research material in terms of cognitive aspects in learning processes (Osinska, Dreszer, Osinski, & Gawarkiewicz, 2013). The main focus in making observation-based conclusions depends not only on age, profession, and interest areas but also on artistic sensibility of active followers. Undoubtedly, visualization stimulates learning about the current state of knowledge. Maps also have an educational value as such applications include interaction mechanisms. The rules of human perception and understanding are applied in designing a visual interface. Particular attention should be paid to how the brain processes color and complex shapes (i.e. fractals). Certainly, color is not an absolute factor on which we can focus the analysis and design of visual message. Because the large individual differences, one should be cautious in anticipating a desired

17

effect. The perception of color changes with age and also depends on cultural patterns of the users (Chen, 2012; Ware, 2004). Modern sophisticated visualizations generated through dimension reduction and mapping algorithms are a kind of graphic layouts allowing for multifaceted and objective data analysis. They have been long used in advanced exploratory analysis and are now an indispensable stage in data mining. Within the existing qualification, visual mining accurately reflects its scientific-empirical objective and interaction with a user, including feedback (Soukup & Davidson, 2002). Nowadays, when marketing is increasingly encroaching on the scientific activity, visualization methods appear to be most appropriate. While designing, one should ensure not only good topological representation but also pay attention to the aesthetical context of visual message. It would seem that this is the area reserved for artists.

Visualization on computer science lessons New graphic forms and animations, as well as appropriately selected visualizations are aimed at tackling teaching problems, especially when we think about teaching of disciplines that are considered difficult, such as computer science. Information Technology is gaining increasing recognition; it is becoming a universal language of almost many disciplines and provides them with tools and development opportunities. The principal objective of educational alphabetization in terms of the ‘3R model’ (reading, ‘riting and ‘rithmetic) nowadays needs to be extended with computer alphabetization called computational thinking, which includes algorithmic thinking, problem-solving, and the skill of programming, applied in all areas of human activity. However, including programming into the educational canon at almost any level of formal and informal education essentially involves devising new methods, which will allow for acquiring the skills of this difficult discipline more easily and at an early age. Once again, what comes to our aid here is visualization, which by combining learning and playing allows students to learn more easily and quickly, making it possible for them to comprehend complex issues. Steve Jobs once said, „Everybody in this country should learn how to program a computer… because it teaches, how to think”. These words have become the motto for carrying out the global movement The Hour of Code vii reaching tens of millions of students in 170+ countries (hourofcode.com/). In 2013, Mark Zuckenberg developed the Angry Birds Hour of Code Game (Rodgers, 2013); he used characters known to the youngest users from the Angry Birds game to teach programming. Increasingly complex commands of a programming language (Java scripts) are acquired in the course of the angry bird’s solving puzzles in order to catch a green pig. The commands of the programming language are written in a simplified form, as a combination of interrelated blocks. This way, the instructions of repetition or decision-making, which are usually hard to understand or write down at the early stages of learning, become user-friendly and can be properly formulated with little effort by anybody. Since December 2013 almost 37 million people from all over the world have participated in the Hour of Code, and new ones still are coming to solve its tasks. The users’ opinions available on the project’s websites confirm the fact that the friendly visualization of a difficult programming language has breached the barriers and made many people, not necessarily the ones with an Information Technology background, believe that they too can program a computer. When solving complex tasks, visualizations may evolve towards a simpler model, which will allow to easily noticing the properties leading to a solution. One of the options is introducing a suitable graph model in order to present the relations between pieces of data. As early as in 1736, Leonhard Eulerviii used this method to solve the problem of bridges over the Pregolya River in Konigsbergix. Modern technology allows observing the satellite image of Konigsberg bridges, for example by using Google Maps, and creating a graph model on its basis (Figure 10), in order to later consider the visualization created by means of such abstract model. On the basis of the graph obtained it is easy to notice when it is possible to cross one bridge exactly once and return to the starting point. The sufficient condition is for each apex to have an even number of edges connected to it.

18

Figure 10. Anna Kwiatkowska. Konigsberg’s bridges on Google maps and its graph model. (© 2014, A. Kwiatkowska and Creative Commons 2014 by Wikipedia. Used with permission) [Figure 10 about here]

E-Learning Visual Solutions E-learning materials should be designed in order to activate students, with particular attention to faceted visual messages, such as interactive maps and illustrations, exploratory statistics (like Gapminder, 2014), and inference on demand. E-learning services designers forget (or do not know) the basic Infovis principle, “focus+context.” The information architecture of e-learning services is usually linear in structure; it has a tabular layout. The central column is dedicated to educational resources like files, links to websites and movies, quizzes and students activities. While processing the material in a classroom, the teacher opens new content sections and students see an extended list. At the end of the course, they see a sequence of sections sorted by topics or according to a calendar. During the course, it is impossible to estimate a quota of redo the covered material according to the whole content – the rule “focus+context” is not present here. If the teacher decides to make all themes visible at once, they will be arranged in a long list, which is not ergonomic for browsing. Students have to scroll through several screens down, which does not help them evaluate the volume of the course content and discourages them from further learning. It would be more comfortable to use non-linear architecture, for example a circle-segmented structure (Figure 11), or just spherical instead of the linear architecture. Figure 11. Veslava Osinska. Conceptual radial model of e-learning course architecture. The topics’ sequence is identified by numbers: 1,2,3,4… (© 2014, V. Osinska. Used with permission). [Figure 11 about here]

19

Cisco Networking Academy x educational program is an example of global online teaching. The Academy supplements the knowledge, which is essential for learning the rules of computer networks’ functioning with visualizations: texts, graphics, and animations. The nonlinear, interactive curriculum provides numerous attractive forms of teaching. The materials, which the Academy provides online include very intuitive and simple visualizations, such as the one from Figure 12, which makes use of the world map in order to demonstrate the role of modern technology in creating online societies, where international borders and physical limitations do not pose a problem. Figure 12. Anna Kwiatkowska. Cisco Networking Academy - networks assist us in our lives”. A sreenshot. Of Cisco e-learning course’s dialog window (© 2014, A. Kwiatkowska. Used with permission). [Figure 12 about here]

Cisco Networking Academy also includes more advanced forms of teaching, created thanks to specialized visualizations. The Packet Tracer program allows for working in a virtual laboratory, where any combination of computers, servers, routers and switches can be created as virtual networks, and which can emulate the functioning of network devices. A visualization of actual solutions along with a simulation of their functioning allows for learning the rules of computer networks’ functioning, analyzing the way of sending information by them, as well as learning even the most complex topologies and network solutions. Figure 13 presents an example of a network topology, which needs to be extended by adding a computer, providing it with a proper IP address, sub network mask, gate, and connecting selected computers to ports.

20

Having done that, one can test the correctness of the proposed solution by turning on a simulation, which entails sending a package of data via the network, and then observing the process. Visualization has in this case been supplemented with a simulation, creating a learning environment based on augmented reality (AR) (Wojciechowski & Cellary, 2010). By using these solutions, students turn from passive recipients to active participants of the teaching/learning process, learning by doing and experimenting. Figure 13. Veslava Osinska. The scheme of a network topology in Packet Tracer virtual laboratory (© 2014, V. Osinska. Used with permission). [Figure 13 about here]

An inventive solution of how to avoid a traditional grid layout in e-learning systems was presented at the World Conference of Computers in Education (http://wcce2013.umk.pl/). An interactive mind map was used as a management tool of didactic resources (Figure 14). This open source plugin on the Moodle platform has been demonstrated in a series of articles (Debska & Sanokowski, 2013). Figure 14. Barbara Debska & Lukasz Sanokowski. Layout of mindmapping module for Moodle. Screenshot of application developed by Debska, Sanokowski (2013). (Creative Commons, 2014) [Figure 14 about here]

21

Rigid linear architecture has a negative effect on e-learning systems, especially on mobile devices. Students’ attention is distracted from the entire structure of document guidelines and they do not recognize the overall task and content orientation. Not only is the role of visual presentations in e-learning underestimated, but also too much passive audio video streaming is used. This does not diminish the importance of video lectures, but it should be noted that the emotional component during on-line transmission is much smaller than during standard lectures (Szelag, Dreszer, Lewandowska, Medygral, Osiński, & Szymaszek, 2010). Using appropriate information architecture on e-learning websites as well as a parallel use of visualization methods can effectively increase it. Rigid information architecture of the e-learning platforms such as Blackboard, Moodle, or Olat should be abandoned in favor of flexible, dynamical, non-linear forms of graphic layout like Prezi. A typical e-learning system based on the appointed sequence of topics is perceived by students as boring, limited, and emotionally attached to a school learning model. The authors propose to apply a dynamical interactive interface, close to the natural spherical vision form that uses a fractal texture (Osinska et al., 2013). A pilot study has been carried out to explore the collection of scientific articles (Figure 15). Figure 15.Veslava Osinska. Spherical visualization interface. Screenshot from a webpagexi. (© 2014, V. Osinska. Used with permission).

22

Non-linearity in the arrangement of e-learning resources was demonstrated in a 2D mapping approach. A tabular structure is rigid; knowledge maps presented in distinct topologies are, in contrast relative to selfsimilar shapes; they resemble fractal structures. Graphic structures based on fractal patterns help students spontaneously focus their attention on difficult topics, and therefore should be frequently used in elearning systems at different education levels (Szelag, Dreszer, Lewandowska, Medygral, Osiński, & Szymaszek, 2010). It should be noted that a tabular meme is firmly rooted in human consciousness: all unknown, unclassified items, events and phenomena are often successfully presented in the form of periodic systems (see the Figure 4). The authors investigated similar and distinct research fields as well as clusters organization by means of obtained graphic patterns. They also analyzed the dynamics of classification due to data series for different publishing periods within a 10-year step. The results show that visualization of classified documents reveals organization of digital library content and allows identifying hierarchical thematic categories.

TRENDS IN VISUALIZATION Natural shapes in perfect visualization One of the ways of activating the ‘first impression’ is to use natural shapes. Typical course books and educational materials are full of Cartesian shapes. Charts, tables, quadrilaterals, and classic shapes are the elements that students directly associate with the educational process, which is not always attractive and interesting. A natural shape should attract their attention. Such behavior has also been observed during the Places and Spaces exhibitions. Daniel Zeller’s (2007) map entitled Hypothetical Model of the Evolution and Structure of Science, presented in Figure 16, has been very popular. Figure 16. Daniel Zeller. Hypothetical Model of the Evolution and Structure of Science. (© 2007, D. Zeller. Used with permission).

23

[Figure 16 about here]

As the author says, “This drawing conceptualizes science as layers of interconnected scientific fields through a stimulating and creative visual language. Starting with the very first scientific thought, science grows outwards in all directions. Each year, another layer is added to the meteor-shaped manifestation of knowledge. New fields emerge (blue), and established fields (brown) merge, split, or die. The cutout reveals a layering of fat years that produce many new papers and slim years in which few papers are added. Each research field corresponds to a tube-shaped object” (Places & Spaces). Why does a seemingly totally unorganized and color-wise unattractive map attract so much attention? The answer should be looked for in the already described properties of the human visual perception system. While seeking an adequate complete visual resonance, fractal forms should be engaged. Fractals are nonCartesian shapes, which were first discovered and used by Benoît Maldenbrot (1977) in 1975. In the next forty years, thanks to the development of computer technologies, their position in many fields of science was established, although the intensive theoretical and experimental study of their properties is still being conducted. Today nobody doubts that the actual natural world is fractal, starting from the micro-scale, i.e. the structure of the smallest biological organisms, through animal and plant worlds, to geological, landscape, and cosmologic structures. Evolution-wise, the human brain must be naturally adjusted to the perception of fractal shapes, since on savannahs and in forests where man evolved and developed, there were only fractal shapes there; there were no quadrilaterals, triangles or straight lines in the natural environment at that time. Therefore, numerous and easily accessible fractal algorithms should be used in graphic designs and book illustrations to add variety or create complete shapes similar to the natural ones. Unfortunately, during school education students are taught that only objects that are ‘completely’ dimensional can exist in the world. Teachers teach about such objects during geometry lessons, and this system of the world description is present throughout the whole education system. It is thus difficult to break the existing schemes and explain, even to specialists, that Cartesian geometry is very important, but is does not describe the whole complexity of the world – because it cannot do so. Nature is more complex than geometry models invented by man; there constantly exist fractal structures in the human brain, coded in neuronal networks, because such structures have surrounded man for millions of years. Fractals are characterized by an incomplete dimension, and thus their symmetry is often broken, which confuses

24

mathematicians but is very useful to neurobiologists. That feature can be creatively used to make suitable graphics for educational materials and to teach. We are used to think that when an object is rotated by 360 degrees, the same image is always received. It is not the case with fractal structures. They emerge in the calculation process in computer memory, and when rotation transformation is applied, a fractal rotated by 360 degrees does not have the same shape as the original element. Figure 17 shows the evolution of the fractal during the rotation transformation. It can be well seen that typical perception properties, which are easily explained by geometrical objects, totally fail here. It is not possible to see similarity of shapes, which can be easily done when rotating a square, for example. Figure 17. Grzegorz Osinski. Fractal structure transformed in rotational symmetry generates completely new shapes. Graphics generated by means of Apophysics. (© 2014, G. Osinski, . Used with permission).

[Figure 17 about here]

25

What is more, a fractal rotated by 360 degrees is slightly different than the original one. A situation where it is not possible to identify inverse transformations is well known to mathematicians. However, in the case of fractal graphics, it constitutes a difficulty while generating new shapes. From a practical point of view, it should be remembered that fractals are usually representations of dynamical processes; therefore, their static representations in the form of images are problematic. Nevertheless, the surrounding reality is not static either, it keeps changing, and photos taken of people or objects are just temporarily caught frames from the continuously changing landscape. Human eyes and brain act the same way. Despite the classic reduction approach to the perception process, which can be best explained by using static images, human perception system is accustomed to ever changing structures. The mystery of fractals that do not return to their original shapes after symmetric transformation is also the mystery of nature. It is worth quoting from a poem by a Noble Prize winner Wisława Szymborska that “nothing can ever happen twice” (Wisława Szymborska, Calling Out to Yeti [Wołanie do Yeti], 1957). Artists and poets can surly understand such difficult issues better when they create art masterpieces or write poems. The language of math and the paradigm of contemporary education are just now discovering and trying to understand phenomena that are already well known to the world of art. While painting, taking photos, or creating illustrations, the process dynamics is being frozen for a moment. However, fractal structures cannot do that, and thus Zeller’s work presented during the Places & Spaces exhibition might have aroused such interest – it was both intriguing and interesting. A shape created by a graphic designer has all basic features of a fractal structure. A correct application of that knowledge will definitely help to create maps with a natural texture and with a greater number of information, and at the same time will be more interesting and analyzed deeper than classic shapes. Since color has a significant impact on perception, we use a carefully controlled palette forcing the brain to draw upon previous experiences or points of reference when viewing art; thus, each viewer has an individual response to a painting. It is not necessary for the viewer to have any knowledge of fractals to make a connection to the presentation. While comparing the shapes presented in Figure 16 and the map representation in Figure 7, similarities can be definitely discerned. Keeping in mind the symmetry breaking described above, a conclusion can be drawn that the structure of network connections presented on the map will not have that characteristic either. It is difficult to predict the development of network; its recovery or an attempt to control it may always fail. Due to fractal properties it is not easy to control natural spontaneous processes. Will such representation, however, be easier for the user to adopt? Will a required feature of analytical image reception develop in the space of the user’s mind? Such questions can be answered in detail only by properly designed experiments based on a large statistical sample. Based on observation, it can already be said that during the study conducted on the participants of the Places & Spaces exhibition, maps of fractal shapes always aroused the greatest interest. The study participants replied in the survey that the maps ‘attracted attention’. They therefore meet the condition of the ‘first impression’, and whether map designers will know how to use it further depends solely on the properly applied data visualization techniques and the placement of data in an applicable context. Obviously, the role of a guide or a teacher in the correct explanation of the map structure cannot be overestimated. It is the guide or the teacher who should explain and lead the student through the further education process. These issues, however, constitute a totally different problem that belongs to the methodology of education. Hopefully, it will be appropriately applied and will facilitate further development of knowledge presentation methods.

3D Visualization – another dimension. Help or hindrance? We should think about whether popular 3D visualization technologies can be applied in map designing. 3D television and cinema have become a standard now. We have become used to the fact that when we wear special glasses, our brain is tricked; we allow this and often like it. The truth is that 3D projections

26

use our natural perception system to create binocular vision and trick our eyes by delivering an image asymmetrically to one eye and then to the other, using various techniques – polarization, chromatic filters or mixed techniques. This technology requires applying complex electronic configurations supported by computer systems. Technology-related difficulties can be avoided, if while creating 3D illusions we use technologies applied by artists such as Manfred Stadler. Being a street art designer, he shows his works in public places, so that viewers can take photos next to them. The 3D effect while viewing his works is achieved thanks to the author’s spatial imagination and prior complex perspective calculations. A graphic artist needs to know at what angle objects should be painted so that they seem three-dimensional when looked at from a specific distance. It can be achieved by using a simple linear perspective with spatial transformation. The illusion of three-dimensionality can be observed when a real 3D object, such as a person, becomes an element of the layout. When the whole is captured in a photo, an ideal imitation of 3D space can be achieved by eliminating the previously described dynamical image analysis effect (Figure 18). Figure 18. Grzegorz Osinski. The kids ‘walking’ on 3D graphics in a shopping mall. (© 2014, G. Osinski. Used with permission). [Figure 18 about here]

Application of those techniques in designing maps and educational materials can be difficult but possible. One can imagine a graphic, which is a projection of subatomic world, for instance, in which students walk around during their physics class. Photos taken by students during such a lesson and shared in social networks will definitely become a significant educational material, perhaps even more interesting than ‘flat’ illustrations in a book. However, it seems that a common application of such a technique is beyond the education system now. The situation looks different with the presentation of data in the form of classic three-dimensional solids. Instead of designing maps on cards, it might be better to create a three-dimensional model. Just like a

27

globe is a three-dimensional map of the Earth, perhaps it will be possible to place a map on the surface of a sphere, which will show a complex analysis of a large dataset. The authors made such an attempt (Figure 19) and presented the results at a science conference – scientists showed interest, but a proper spatial orientation became a problem (Osinska et al., 2013). It turned out that not all could naturally identify the North Pole as the top axis and the South Pole as the bottom axis. Has the widespread of the GPS, which changes the map orientation based on the movement direction, created a stereotype in our brain that our location is identified against our position? Figure 19. Veslava Osinska. 3D Data globe versus information map – comparison of output topology of data representation. (© 2014, V. Osinska. Used with permission). [Figure 18 about here]

Have we forgotten which direction a compass shows and how a map should be oriented? Probably yes; however, the P&S exhibition also shows data mapped onto the surface of spheres and they are very popular among students. Perhaps thanks to having contact with presentations in the form of a sphere that shows an actual shape of our planet, students will not have problems with spatial orientation and will remember what the term north direction and its representation on a flat cartographic map really mean.

Neuroaesthetics Aspects When we read about Rembrandt who “loved what he painted and only painted what he loved,” we can understand an emotional level of perception strictly compared with creative processes. Sometimes visualization starts to play the same role as an artistic presentation of large structural data. The users of a visual message should show behavior similar to the behavior of the art gallery visitors. A desired effect of communication would be achieved when the user is deeply engaged during the visual perception thanks to his or her aesthetic sensitivity. It has already been emphasized that the state of the brain is indivisible – phonology and semantics are not separate perception systems. They operate at the same time and are mutually coupled. In order to create visual communication in the right way, it has to be understood how current neuroscience attempts at describing the issue of aesthetic experience. This problem belongs to neuroaesthetics, which studies creative processes in art and tries to understand the mechanism of the human brain during such processes. Neuroaesthetics is not just the study of artistic experiences, but it also emphasizes the crucial influence of the brain study on the understanding of human nature. The central

28

point of nthe euroaesthetics is the study of perception laws, which art creation is subjected to, both at the level of creating and viewing. Studies within neuroaethetics have been conducted for years; the beginning of the scientific development of neuroaesthetics is arbitrary dated to the work of Semir Zeki (1999). They are interdisciplinary studies; however, a consensus accepted by specialists from different disciplines is difficult. Currently, there is an ongoing discussion among specialists regarding issues related to neuroarthistory. The pioneering work by John Onians (2008) has opened an utterly new chapter in the study related to interaction with the beauty in a visual communication. Hopefully, the work continuing in a growing circle of scientists, especially art historians, book historians, and neuroaestheticians, will help to discover unknown dependencies and historic implications of creating universal communication. First attempts at a precise definition of art from the perspective of neuroscience were made by Ramachandram in his already classical works (Ramachadran & Hirstein, 1999; Ramahadran, 2010). Following his directions, especially the ‘Ramachandran’s nine laws of aesthetics,’ it is possible to create a visual communication, which follows classical rules of aesthetics. However, the sole principles of creating an aesthetic communication do not yet explain why visual artistic impressions are experienced in such a way. In recent years, new studies results have been appearing, which justify the proposed laws of aesthetics and thus show the complexity of the problem. Perhaps it is the fractal structure, which shows how the brain creates images in the space of the mind based on visual impressions. This dynamical way is generated by perception experiences in the limbic system, which can be defined as ‘aesthetic impressions.’ However, it should not be underestimated that one of the oldest structures of the brain, i.e. the limbic system reacts to danger directly, omitting the cortex paths. That mechanism, which has developed evolutionally, allows avoiding danger. However, it cannot be said that its role is meaningless in the analysis of images. In the limbic system of the brain, information couples with the system of emotions. This path is probably responsible for aesthetic experiences of the viewed image. Before visual information reaches neuronal structures located in the neocortex, where analytical processes of information processing will take place, it will surely first activate the structures of the limbic system. This will result in a specific emotional state, from the perspective of which a further analysis will be conducted. Thus, special attention should always be paid to maintaining a specific balance between the subject-related content of the communication and its aesthetic universality.

SUMMARY AND CONCLUSION The authors show that visualization, in parallel with popularity in the net applications, is more and more useful in learning.Authors concentrate on the role of maps as teaching tools during the education process. At the beginning of the chapter content map (Figure 1) aims to introduce the readers visually to the discussed issues and problems of visualization, considering knowledge communication through the information maps, their practical application and e-learning architecture conception as well as cognitive and perceptual, technological and contemporary art aspects of Infoviz. The process of map reading can be divided into layers (Figure 3), each of which is responsible for the components of the analysis: image characteristics grouping, shape recognition, third dimension recognition, and assembly of objects into the final scheme. Parallel identification of the objects in the mind influences long-term memory (LTM), the development of semantic terms, and the process of creation of new ideas. In many works on visualization, researchers refer to the visual perception model directly resulting from the Gestalt principles. We believe that the rules defined by psychologists in the 1930s are now being overused, especially in visualizations and infographics, or wrongly interpreted. One should see that during the visual analysis not only shape is significant, but also the structure of the viewed object or scene.

29

Tabular shapes dominate in information architecture. This is the structure of website layouts. E-learning courses also use this traditional convention. The alternatives are non-linear layouts, e.g. network or fractal. Mind maps have non-linear architecture in principle and therefore, implementation of such tools in e-learning platforms should be supported and developed. Special attention has been given to 3D visualization, supported with interesting examples, including the discussion on its advantages and disadvantages in the correct interpretation of images. Fractal structures are exceptionally intuitive in perception and reception because they originate from (or resemble) nature. This explains why fractal-like visualizations are perceived first. Visual communication messages should be constructed following such patterns. This typical neuroscience issue has been solved in nature by fractal structures – easy to compute iteratively, but reflecting the structural complexity in the form of aesthetic communication. The last one is the subject of neuroaesthetics - a cross-disciplinary research field related to art, history of art, cognitive and computer sciences, communication, and mathematics (Zeki, 1998, 1999; Onians, 2008). It is not possible to mention all issues or troubles of visualization, which take place in communication and learning processes today. This chapter attempts to join the usable aspects of visualization that currently afflict both Infoviz researchers and practitioners. Data and/or information visualization became an interdisciplinary methodology. The authors are from different work backgrounds and majored in different subjects, but their specializations intersect in visualization space. Therefore we only present those studies, which according to the authors are most significant in their area.

ACKNOWLEDGEMENTS The authors wish to thank Wlodzislaw Duch for inspiration and practical suggestions.

REFERENCES Alivasatos, P. (2002). The Brain Activity Map Project and the Challenge of Functional Connectomics. Neuron 74(6). Allport, F. H. (1955). Theories of perception and the concept of structure: A review and critical analysis with an introduction to a dynamic-structural theory of behavior. NJ, US: John Wiley & Sons Inc. Barres, V., & Lee, J. (2014). Template Construction Grammar: From Visual Scene Description to Language Comprehension and Agrammatism, Neuroinformatics 12(1), 181-208. doi: 10.1007/s12021-013-9197-y. Börner, K., Klavans, R., Patek, M., Zoss, A. M., Biberstine, J. R., Light, R. P., Lariviere, V. & Boyack, K. W. (2012). Design and update of a classification system: The UCSD map of science. PLoS ONE, 7, e39464. Retrieved September 18, 2014, from: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0039464 Chen, C.H. (2012). Emerging Topics in Computer Vision and its Application. Word Scientific. Cossart, R., Aronov, D., & Yuste, R. (2003). Attractor dynamics of network UP states in the neocortex. Nature 423:283-8.

30

Debska, B. & Sanokowski, L. (2013). Automatic Generation of Mindmaps in Courses Implemented on Moodle Platform. In N. Reynolds, M. Webb, in coop. with V. Dagiene, M. Sysło (Ed.) Proceedings of 10th IFIP World Conference of Computers in Education, Torun, Poland, July 1-7, 2013, pp. 72-80. Duch, W. (2007a). Creativity and the Brain. In A.-G. Tan (Ed.) A Handbook of Creativity for Teachers (pp. 507-530). Singapore: World Scientific Publishing. Duch, W. (2007b). Intuition, insight, imagination and creativity. In IEEE Computational Intelligence Magazine, 2(3), 40-52. Duch, W. (1998). Platonic model of mind as an approximation to neurodynamics. In: ed. S-i. Amari, S-I., & Kasabov, N. (Eds.), Brain-like computing and intelligent information systems, Singapore: Springer, pp. 491-512. Duch, W., & Diercksen, G. (1995). Feature Space Mapping as a universal adaptive system. Computer Physics Communications, 87(3), 341-371. Duch, W. & Grudzinski, K. (2001). Prototype based rules – a new way to understand the data. In IEEE Proceedings Neural Networks Vol. 3. (pp. 1858-1863). IEEE. Dürsteler, J. C. (2007). Diagrams for Visualization. The digital magazine of InfoVis.net. Retrieved May 31 2014 from: http://www.infovis.net/printMag.php?num= 186&lang=2. Few, S. (2009). Now You See It: Simple Visualization Techniques for Quantitative Analysis. Oakland, CA: Analytics Press. Gapminder (2014). Gapminder for fact-based world view. Retrieved September 18, 2014, from: http://www.gapminder.org/ Klavans, R. & Boyack, K. W. (2006). Quantitative Evaluation of Large Maps of Science. Scientometrics 68(3), 475-499. Klavans, R. & Boyack, K. W. (2007). Maps of Science: Forecasting Large Trends in Science. In K. Börner and J. M. Davis (Eds.) 3rd Iteration: The Power of Forecasts, Places & Spaces: Mapping Science. Retrieved May 31 2014 from: http://scimaps.org/maps/map/maps_of_science_fore_50/ Kuhn, T. S. (1996). The Structure of Scientific Revolutions. 3rd edition. Chicago and London: University of Chicago Press. Lengler, R. & Eppler, M. J. (2007). Towards a periodic table of visualization methods of management. In GVE '07 Proceedings of the IASTED International Conference on Graphics and Visualization in Engineering (pp. 83-88). Anaheim, CA: ACTA Press. Luther, J., Kelly, M., & Beagle, D. (2005). Visualize This. Library Journal, 3(1). Retrieved May 31 2014 from: http://www.libraryjournal.com/article/ CA504640.html. Lynch, P. J., & Horton, S. (2008). Web Style Guide: Basic Design Principles for Creating Web Sites. 3rd edition. Yale University Press. Mandelbrot, B. (1977). Fractals: Form, Chance and Dimension. W. H. Freeman and Co.

31

McCormick, B. H., DeFanti, T. A., & Brown, M. D. (1987). Computer Graphics. In ACM SIGGRAPH 21(6). Retrieved September 18, 2014, from http://www.evl.uic.edu/core.php?mod=4&type=3&indi=348 Necka, E., Orzechowski, J., Szymura B. (2006). Cognitive psychology, Warszawa, Poland: PWN (in polish). Onians, J. (2008). Neuroarthistory: From Aristotle and Pliny to Baxandall and Zeki. New Haven, CT: Yale University Press. Osinska, V., Dreszer, J., Osinski, G., Gawarkiewicz, M. (2013). Cognitive Approach to Classification Visualization. End Users Study. In Slavic, A., Akdag Salah, A., & Davies, S. (Eds.). Classification & visualization: Interfaces to knowledge. Proceedings of the International UDC Seminar 24 – 25 October 2013, (pp. 273-281). Würzburg: Ergon Verlag. PISA 2009 Technical Report. OECD Publishing. ISBN 978-92-64-04018-2. Retrieved September 17, 2014, from: www.oecd.org/pisa/pisaproducts/50036771.pdf. Rafols, I., Porter, A. L., & Leydesdorff, L. (2010). Science Overlay Maps: Anew Tool for Research Policy and Library Management. Journal of the American Society for Information Science and Technology, 61(9), 1871-1887. Ramachandran, V. S. & Hirstein, W. (1999). The Science of Art. A Neurological Theory of Aesthetic Experience, Journal of Consciousness Studies 6(6-7), 15–51. Ramachandran, V. S. (2010). The Tell-Tale Brain: A Neuroscientist's Quest for What Makes Us Human. W. W. Norton & Company. Rodgers, M. (2013). Hour of Code Angry Birds Code Game. Retrieved September 19, 2014 from: https://www.youtube.com/watch?v=tIVVelpRhTA Shneiderman, B. (2009). Treemaps for space-constrained visualization of hierarchies. Retrieved May 31 2014 from: http://www.cs.umd.edu/hcil/treemap-history/. Soukup, T. & Davidson, I. (2002). Visual Data Mining: Techniques and Tools for Data Visualization and Mining. 1st edition. Wiley. Szelag, E., Dreszer, J., Lewandowska, M., Medygral, J., Osiński, G., Szymaszek, A. (2010). Time and Cognition from the Aging Brain Perspective: Individual Differences, Eliot Werner Publication. Ware, C. (2004). Information Visualization. Perception for Design. San Francisco, CA: Morgan Kaufmann. Wojciechowski, R. & Cellary, W. (2010). Interactive learning environment based on augmented reality. Edu@kcja, Magazyn Edukacji Elektronicznej 1, 42-48 (In polish). Zeki, S. (1998). Art and the Brain. The Brain 127(2). Zeki, S. (1999). Inner Vision. Oxford University Press.

32

Zeller, D. 2007. Hypothetical Model of the Evolution and Structure of Science. In K. Börner & J. M. Davis (Ed.) 3rd Iteration: The Power of Forecasts, Places & Spaces: Mapping Science. Retrieved May 31 2014 from: http://scimaps.org.

ADDITIONAL READING Blondel, V., Siguret, C., & Pina, J. (2006). 100 CV pour gagner! Marabout. Börner, K. & Polley, D. E. (2014). Visual Insigths. A practical guide to Making Sense of Data. Cambridge, MA: The MIIT Press. Börner, K. (2010). Atlas of Science. Cambridge, MA: The MIT Press. Burkard, F.P., Wiedmann, F. & Kunzmann, P. (1999). DTV-Atlas Philosophie. Warszawa, Poland: Proszynski i S-ka (in polish). Burke, J. (1985). The Day the Universe Changed. London, UK: The London Writers. Burke, J. (1999). The Knowledge Web. London, UK: The London Writers. Cairo, A. (2013). The Functional Art. An Introduction to Information Graphics and Visualization. Berkeley, CA: New Riders. Chen, Ch. (2006). Information visualization: Beyond the Horizon. 2nd edition. Springer Science & Business Corbalán, F. (2012). La proporción áurea. El lenguaje matemático de la belleza. Barcelona: Spain: RBA Contenidos Editoriales y Audiovisuales, S.A. Emerging Topics in Computer Vision. (2012). Ch. Chen (Ed.). Singapore: World Scientific Publishing Co. Glass, L. & Mackey, M. C. (1988). From Clocks to Chaos. The Rhythms of Life. Princeton, New Jersey: Princeton University Press. Kalbach, J. (2007). Designing Web Navigation. O'Reilly Media. Larsen, R. (2010). The Selected World of T. S. Spivet. Penguin Books. Marcus, G. F. (2001). The Algebraic Mind. Massachusetts Institute of Technology. McCandless, D. (2009). The Visual miscellaneum. A colorful Guide to the Rold’s Most Consequential Trivia. New York, NY: HarperCollins Publishers. Morville, P. & Callender, J. (2010). Search Patterns. O'Reilly Media. Morville, P. & Rosenfeld, L. (2006). Information Architecture for the World Wide Web. 3rd Edition. O'Reilly Media. Nielsen, J. & Tahir, M. (2001). Homepage Usability: 50 Websites Deconstructed. New Riders Publishing.

33

Osinska, V. & Bala, P. (2010). New Methods for Visualization and Improvement of Classification Schemes : the Case of Computer Science. Knowledge Organisation 37(3), 157-172. Osinska, V. (2010). Documents Visualization and Retrieval. Warszawa: Poland: SBP (in polish). Popek, S. (2003). Colours and Psyche. Lublin: Poland: WUMCS (in polish). Schroeder, W., Martin, K., & Lorensen, B. (2004). The Visualization Toolkit. 3rd edition. Kitware Inc. Steele, J. & Iliinsky, N. (2010). Beautiful Visualization. Looking at Data Through the Eyes of Experts. Canada: O’ Reilly. Stewart, I. (2007). Why Beauty is Truth. Joan Enterprises. Stewart, I. & Cohen, J. (1997). Figments of Reality. Warszawa, Poland: Proszynski i S-ka (in polish). Towards A Unified Theory of Development. Connectionism and Dynamic Systems Theory Re-considered. (2009). J. P. Spencer, S. C. Thomas & J. L. McClelland (Eds.). Oxford, UK: Oxford University Press.

34

KEY TERMS AND DEFINITIONS Attractor: is a general activity function that describes big and complex sets of dynamical particles. In cognitive science the activity of neural correlates can be identified by attractor systems. The attractor is defined as a smallest unit that cannot be decomposed. General activity of a whole brain may be described as a system of multiple attractors. Cartesian shape – an integer dimension of a classical geometric shape such as square, triangle, or circle, in opposition to a non-Cartesian shape with non-integer dimension. Graph: Mathematical construction, which consists of the set of nodes and edges. Nodes usually represent investigated objects, while edges – the links (relations) between them. Graph is popular visualization method in many areas like science, business, medicine, or education. Infovis: Information Visualization – the abbreviation is commonly used in professional literature on visualization. Neural correlates: this is a minimal set of general neuronal events and mechanisms sufficient for a specific activity of the brain. In visual systems we have correlates responsible for perception, idea creation, and other cognitive activities. In neuronal systems we have both type of transfers: information about visual perception and energy of neural correlates. Nonlinearity: a feature of a system or a structure where output is not directly proportional to input – just like in perception of human visual systems. In general it is practical consequence of Aristotle statement: “The whole is greater than the sum of its parts.” Preconceptions: They are intuitive and preconceived ideas about processes and objects that we have never reflected upon, but which have appeared spontaneously during learning or another activity. Resonance: is a very general phenomenon that happens in many natural and artificial systems. The general sense represents that the energy or information in a system exchanges from one form into another at a particular rate. Science map: Graphical representation of scientific domains using bibliographic, bibliometric, and scientomeric data. Science maps reveal the domain structure of science(s) and collaboration ties between researchers such as co-authorship, co-citations. Eugene Garfield (1994, http://wokinfo.com/essays/scientography-mapping-science/) called these representations scientographs. i

Places & Spaces. Mapping Science. Online exhibition of science maps at: http://scimaps.org

ii

Many Eyes is a free site to upload, visualize, discuss, and share visualization datasets and results: http://www958.ibm.com/software/analytics/manyeyes/ and http://www.bewitched.com/manyeyes.html iii

Gephi is a free open source interactive visualization and exploration platform for all kinds of networks and complex systems. The webpage: www.gephi.org iv

The Periodic Table of Visualization Methods – an interactive online application showing visualization techniques with examples: http://www.visual-literacy.org/periodic_table/periodic_table.html v

Gapminder is an open source software for statistical data visualization and animation: www.gapmnder.org

35

vi

Map of Science, available at: www.mapsofscience.com, is web service developed by SciTechStrategies and dedicated to a specific kind of visualization – the UCSD maps. vii

The Hour of Code is a web service dedicated to develop programming skills of students: http://code.org

viii

Leonhard Euler (1707-1783), Swiss mathematician and physicist is considered the founder of the graph theory.

ix

Konigsberg in 1736 belonged to Prussia. At present it is called Kaliningrad and belongs to Russia.

x

Cisco Networking Academy Program helps students develop the foundational ICT skills needed to design, build, and manage networks. The website at: http://cisco.netacad.net xi

Authors’ online application to explore computer science thematic categories and their dynamics: http://wwwusers.mat.umk.pl/~garfi/vis2009v3/