Dream Cartography

Research Collection

Doctoral Thesis

Dream Cartography. Mapping Dream Space and Content for an Exploratory Analysis of Dreams

Author(s): Iosifescu Enescu, Cristina M.

Publication Date: 2019

Permanent Link: https://doi.org/10.3929/ethz-b-000381774

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.

ETH Library DISS. ETH NO. 26164

Dream Cartography Mapping Dream Space and Content for an Exploratory Analysis of Dreams

A thesis submitted to attain the degree of DOCTOR OF SCIENCES of ETH ZURICH (Dr. sc. ETH Zurich)

presented by

Cristina Mihaela IOSIFESCU ENESCU

Dipl.-Eng. Geodesy, UTCB Bucharest MSc. Psychology, UZH Zurich

born on 30.10.1979

citizen of Zurich ZH and Romania

accepted on the recommendation of

Prof. Dr. Lorenz Hurni Prof. Dr. William Cartwright

2019

For my dear husband Ionuț and for our three sons

Row, row, row your boat Gently down the stream Merrily, merrily, merrily, merrily Life is but a dream

Nursery rhyme, 1852

Summary The objective of Dream Cartography is to map dream content by developing new visualization methods and adapting existing ones as graphical means for an exploratory dream analysis. It encompasses documenting, modelling, and graphically depicting the dream space and other relevant psychological aspects of dreams in a meaningful, abstracted way.

The dream space is fundamentally different from the real space, especially because it is not possible to consciously visit the place of the dream again. One way to address this setting issue is through data acquisition, and therefore means for getting a more accurate description of the dream experience and particularly of the dream setting are examined. Furthermore, a new type of Volunteered Geographical Information (VGI) is proposed: the fictional VGI (F-VGI). The social aspect of dreams is exploited, arguing that not only the dreamer, but also a third party may contribute to the fictional geography.

The dream setting is modelled, based on the systematic study of dream reports, on literature in dream psychology, geography and narratology, and on two empirical surveys. The first one is an online survey, with a convenience sample, answered by more than 300 persons and deals with general aspects of dream settings, such as what are elements, by which dreamers recognize a dream place. The second one is a paper-and-pencil survey, applied to an academic group composed of 30 cartography and geoinformation specialists. It contributes to the definition of the components of a setting. For example, one component is the dreamer’s familiarity with that place. This familiarity is used for building a straightforward model, which is the base for the place cookie diagram, visualized in form of concentric circles. In total, 26 dimensions are identified, which extensively characterize a setting. These dimensions are grouped into eight factors, which are: relation to geospace vs. role of space, society vs. morphology, individual vs. general exposure, and attention vs. environment. The obtained model is used to create a visual profile of the dream setting, named setting spider, in form of a radar chart with eight “legs”. Moreover, a similar model is built for a dream scene, resulting into the so-called event spider. The proposed profiles enable comparisons of different dream settings, or of different dreams, respectively. They can be used as uni- or multivariate point symbols on a thematic layer on top of a dreamland map. Additionally, the set-up of Web platform for mapping dreams is discussed and a functional prototype is implemented, featuring tools for data acquisition, for analyzing, graphical tools related to map creation and editing (e.g. map collage), and visualizations for different dream topics, such as picture frames for emotions or a graph-network for social interactions.

Finally, it is argued that the developed data models can be also adopted for real-life settings and events, therefore contributing to a better awareness of space in general, and acknowledging personal experiences as valuable for mapping real-life places. In a future continuation of this research, the application of such a dynamical model of the setting to real-life places could for example bring new insights for touristic places, or for spatial and settlement development in an urban environment.

ii Zusammenfassung (German) Das Ziel der Traum-Kartografie ist, Trauminhalte abzubilden, indem neue Visualisierungsme- thoden entwickelt und bestehende als grafisches Mittel für eine explorative Traumanalyse an- gepasst werden. Sie umfasst das Dokumentieren, Modellieren und grafische Darstellen der Traum-Umgebung und anderer relevanten psychologischen Aspekten von Träumen in sinnvol- ler, abstrahierter Weise.

Der Traum-Raum unterscheidet sich grundlegend vom realen Raum, zumal es nicht möglich ist, den Ort eines Traumes wieder bewusst zu besuchen. Eine Möglichkeit, dieses ortsbezogene Problem zu lösen, besteht in der Datenerfassung. Daher werden Mittel untersucht, um eine ge- nauere Beschreibung der Traumerfahrung und insbesondere des Traumortes zu erhalten. Aus- serdem wird ein neuer Typ von Volunteered Geographical Information (VGI) vorgeschlagen: die fiktionale VGI (F-VGI). Der soziale Aspekt der Träume wird damit berücksichtigt und es wird argumentiert, dass nicht nur Träumer, sondern auch Dritte zur fiktiven Geografie beitragen können.

Der Schauplatz, die Umgebung des Traums, wird modelliert - dies basierend auf systematischer Untersuchung von Traumberichten, Literatur zu Traumpsychologie, Geografie und Narratolo- gie und auf zwei empirischen Umfragen. Die erste ist eine Online-Umfrage mit einer willkür- lichen Stichprobe von mehr als 300 Personen, die sich mit allgemeinen Aspekten von Traum- Schauplätzen befasst, z. B. mit Elementen, anhand derer Träumer einen Traumort erkennen. Die zweite ist eine Papier-und-Bleistift-Umfrage, die mit einer Gruppe von 30 akademischen Kartografie- und Geoinformations-Experten durchgeführt wurde. Diese trägt zur Definition der Bestandteile der Umgebung bei. Eine Komponente ist beispielsweise die Vertrautheit des Träu- mers mit dem Traumort. Diese Vertrautheit wird verwendet, um ein einfaches Modell zu erstellen, das die Grundlage des Place-Cookie-Diagramms bildet, ein Diagramm in Form konzentri- scher Kreise. Insgesamt werden 26 Dimensionen identifiziert, die den Schauplatz weitgehend charakterisieren. Diese Dimensionen werden in acht Faktoren gruppiert: Bezug zum geogra- phischen Raum und Rolle des Raums, Gesellschaft und Morphologie, individuelle und allge- meine Abdeckung sowie Aufmerksamkeit und Umwelt. Das erhaltene Modell wird verwendet, um ein visuelles Profil der Traumumgebung zu erstellen, den Setting Spider in Form einer Ra- darkarte mit acht „Beinen“. Dazu wird ein ähnliches Modell für eine Traumszene entwickelt, aus dem der sogenannte Event Spider hervorgeht. Die vorgeschlagenen Profile ermöglichen den Vergleich verschiedener Traumschauplätze bzw. verschiedener Träume. Sie können als uni- oder multivariate Punktsymbole in einer thematischen Ebene auf einer Traumlandkarte verwendet werden. Darüber hinaus wird der Aufbau eines Web-Portals zur Kartierung von Träumen erörtert und ein funktionaler Prototyp implementiert, der Tools zur Datenerfassung, zur Analyse, grafische Werzeuge zur Kartenerstellung und -bearbeitung (z. B. Kartencollage) und Visualisierungen für verschiedene Traumthemen enthält, wie z. B. Bilderrahmen für Ge- fühle oder Graphen-Netze für soziale Interaktionen.

iii Schliesslich wird argumentiert, dass die entwickelten Datenmodelle auch für reale Umgebun- gen und Ereignisse übernommen werden können, was zu einem besseren Raumbewusstsein im Allgemeinen beitragen könnte und persönliche Erfahrungen als wertvoll für die Kartierung re- aler Orte anerkennt. In einer zukünftigen Fortsetzung dieser Forschung könnte die Anwendung eines solchen dynamischen Modells der Umgebung auf reale Orte neue Erkenntnisse bringen, die für touristische Orte oder für die Raum- und Siedlungsentwicklung im städtischen Umfeld relevant sein können.

iv Rezumat (Romanian) Obiectivul Cartografiei Viselor este de a carta conținutul viselor prin dezvoltarea de noi metode de vizualizare, precum și adaptarea celor existente, ca mijloace grafice pentru o analiză explorativă a viselor. Acesta cuprinde documentarea, modelarea și reprezentarea grafică atât a spațiului visat, cât și a altor aspecte psihologice relevante pentru vise, într-un mod compact, abstractizat.

Spațiul visat diferă în mod fundamental față de spațiul real, mai ales pentru că nu este posibil să revizităm locul din vis în mod conștient. O modalitate de a aborda această problemă a locului visat este prin achiziția de date și, prin urmare, sunt examinate diferite mijloace pentru a obține o descriere cât mai exactă a experienței visului și în special a locurilor din vis. Mai mult, este propus un nou tip de informație geografică voluntară (VGI): VGI fictiv (F-VGI). În acest fel este exploatat aspectul social al viselor, argumentând că nu numai cel care a visat, ci și o terță parte, poate contribui la geografia fictivă.

Locul, ambianța din vis, sunt modelate cu ajutorul unui studiu sistematic al rapoartelor despre vise, al literaturii de specialitate în psihologia viselor, în geografie și naratologie și cu ajutorul a două sondaje empirice. Primul este un sondaj online, cu un eșantion de conveniență, la care au răspuns peste 300 de persoane. Acest sondaj cuprinde aspecte generale ale locurilor din vis, cum ar fi elementele prin se recunoaște un loc în vis. Al doilea este un chestionar de tip creion- și-hârtie, aplicat unui grup de 30 de experți în cartografie și geoinformație. Acesta contribuie la identificarea componentelor ce definesc un loc în percepția subiectivă. De exemplu, una dintre componente este familiaritatea celui care visează cu acel loc. Familiaritatea este utilizată pentru a construi un model simplu, care stă la baza diagramei „Place Cookie”, sub formă de cercuri concentrice. În total, sunt identificate 26 de dimensiuni, care caracterizează în mod complet un loc. Aceste dimensiuni sunt grupate în opt factori: relația cu spațiul geografic și rolul spațiului, societatea și morfologia, expunerea individuală și cea generală, respectiv atenția și mediul înconjurător. Modelul obținut este folosit pentru a crea un profil vizual al locului subiectiv din vis, numit Setting Spider, sub forma unei diagrame de tip radar ca un "păianjen" cu opt "picioare". În continuare, un model similar este construit și pentru o scenă din vis, rezultând așa-numitul Event Spider. Profilurile propuse permit compararea diferitelor locuri subiective din vis, respectiv a diferitelor vise. Ele pot fi folosite ca simboluri punctiforme univariate sau multivariate, într-un strat tematic pe o hartă a viselor. În continuare sunt discutate cerințele pentru o platformă de cartare a viselor pe internet, fiind implementat și un prototip funcțional de platformă. Platforma de cartare a viselor pe internet conține instrumente pentru achiziția de date, pentru analiză, instrumente grafice legate de crearea și editarea hărților (de exemplu colaje) și vizualizări pentru diferite teme din vis (cum ar fi rame foto pentru reprezentarea emoțiilor sau o rețea de tip graf pentru interacțiunea socială).

În final se afirmă că modelarea propusă ar putea fi aplicată și pentru locurile și evenimentele din viața reală, contribuind astfel la o mai bună conștientizare a spațiului în general și la

v recunoașterea valorii experiențelor personale pentru crearea unei hărți. În viitor, acest model dinamic al locurilor subiective s-ar putea aplica și pentru zone turistice sau pentru diverse zone rezidențiale din viața reală, ceea ce ar putea, de exemplu, să aducă perspective noi în industria turistică sau în dezvoltarea teritorială în mediul urban.

vi Table of Contents

Summary ...... i

Zusammenfassung (German) ...... iii

Rezumat (Romanian) ...... v

List of figures ...... xi

List of tables ...... xv

List of scientific publications ...... xvii

Chapter 1. Introduction ...... 1

Motivation ...... 2 Problem Statement and Research Questions ...... 3 Methodology ...... 5 Overview of the Research Articles ...... 6 Relevance for Science and Society ...... 7 Structure of the Thesis ...... 8

Chapter 2. Background ...... 9

Different Geographies ...... 10 Human Space ...... 11 Dream Research ...... 13 The Web as a Communication Medium ...... 15 Contribution to Research ...... 15

Chapter 3. Dream Cartography Concepts ...... 17

Paper I. Toward Dream Cartography: Mapping Dream Space and Content .. 18 1 Introduction ...... 18 2 Setting as a Basic Construct in Dream Cartography ...... 20 3 Dream Research: a Brief Overview ...... 21 4 Integrating Dream Cartography into Current Cartographic Research ...... 21 5 Mapping Dreams: Workflow ...... 23 6 Visualization Examples ...... 29 7 Conclusion and Future Work ...... 36

vii 8 References ...... 37 Paper II. Fictional volunteered geographic information in Dream Cartography ...... 40 1 Introduction ...... 40 2 State of the art ...... 41 3 A study on dream locations ...... 47 4 Conclusions and outlook ...... 50 5 References ...... 50 Paper III. Place Cookies and Setting Spiders in Dream Cartography ...... 54 1 Introduction ...... 54 2 State of the Art ...... 55 3 Methods ...... 60 4 Data Models ...... 64 5 Application in Dream Cartography ...... 71 6 Summary and Discussion ...... 75 7 Bibliography ...... 77

Chapter 4. Proof-of-Concept Implementation ...... 81

Paper IV. Cartographic Tools for Mapping Dreams ...... 82 1 Introduction ...... 82 2 Background ...... 83 3 Requirements for the Web Platform for Mapping Dreams ...... 84 4 Initial Design of the User Interface ...... 88 5 A Functional Prototype ...... 90 6 Application Example: Use Case ...... 93 7 Conclusions and Outlook ...... 94 8 References ...... 95

Chapter 5. Concluding Remarks ...... 97

Synopsis of Results ...... 98 Discussion ...... 101 Outlook ...... 102

References ...... 104

Appendices ...... 107

Appendix A. Survey for setting profile ...... 108 Appendix B. Statements and their values for the Setting Profile ...... 112

viii Appendix C. Statements and their values for the Event Profile ...... 116

Acknowledgments ...... 118

Curriculum vitae ...... 120

x List of figures

Chapter 2 Figure 1 Dream settings, subdivided according to their familiarity; from (Strauch & Meier, 1996, p. 108) ______15

Chapter 3, Paper I Figure 1 Morphing example of cities of Göttingen and Zurich, which appear as a condensed place in a dream; Original images: Bing Maps, Bird’s Eye ______30

Figure 2 Dreamer's living locations in time; in red the “Iron Curtain”, which separated East from the West Germany before 1990; Data: Google Maps & scribblemaps.com __ 31

Figure 5 Example for the depiction of the social interaction network; “I” stands for the dreamer’s person ______35

Figure 6 Proposed aging masculine profiles to be used for characters in the social interactions network of a dream ______36

Chapter 3, Paper II Figure 1 Distribution of memories and dreams in terms of a person’s age at the time of the original experience; after Grenier et al (2005, p. 286). ______45

Figure 2 The schema, on how a morphed city was composed from two city plans, of Zurich (top) and Göttingen (bottom); Original source data: OpenStreetMaps contributors 46

Figure 3 The map of the fictional territory Westeros (left) compared with the map of Ireland (right); Sources: http://quartermaester.info/ and Google Maps ______47

xi Figure 4 Dream Location Questionnaire, page with general questions on dream location. __ 48

Chapter 3, Paper III Figure 1 Visually summarized uses of concentric circles for classifying the social network of a person. (A) Hierarchically inclusive levels of acquaintanceship by Dunbar (2010); (B) Circles of support by Falvey et al. (1997) ______59

Figure 2 Smart-spider diagram for Swiss elections 2009 (Hermann, 2010) ______60

Figure 3 Example of a filled place cookie ______66

Figure 4 The setting spider or place metadata profile: defining factors for a subjective perception of a setting and their components ______68

Figure 5 The dream or event spider, metadata profile for a dream / an event ______70

Figure 6 Map using the place cookie as point symbol for spatial locations, Background data: OpenStreetMaps contributors ______72

Figure 7 Personal dreamland with dream setting spiders on it ______73

Figure 8 Close-up of personal dreamland, a dream from northern Italy ______74

Chapter 4, Paper IV Figure 1 Predefined styles for GIS data visualization in a Web application (Probst, 2013) __ 86

Figure 2 Word cloud generated with Voyant Tools ______87

Figure 3 Place cookie summarizing the dreams in a dream series based on the dream setting familiarity (1-6) and showing also the emotional valence of the dream (+/-/=) ___ 88

Figure 4 Initial proof-of-concept mock-up for the design of the user interface of the Web platform for mapping dreams ______89

Figure 5 Dream / Event Spider Profile of the selected dream, generated on the Web platform for mapping dreams ______92

Figure 6 Setting Spider Profile of the selected dream, generated on the Web platform for mapping dreams ______92

xii Figure 7 Example of a dream map created with the interactive Web platform for mapping dreams ______93

xiii

xiv List of tables

Chapter 1 Table 1 Research questions covered by the publications ______4

Chapter 2 Table 2 Differences between abstract and human space from Tilley (Tilley, 1994, p. 8) ___ 12

Table 3 Coding dream settings as proposed by the Hall /Van de Castle coding system ____ 13

Chapter 3, Paper I Table 1 Instructions for keeping a dream log ______24

Table 2 Questionnaire for a dream scene ______25

Table 3 Criteria proposed for dream visualization and considerations made toward their modeling into parameters ______26

Chapter 3, Paper II Table 1 Examples of dream locations given by people in the dream location survey ______49

Chapter 3, Paper III Table 1 Overview of the survey ______62

Table 2 Ordinal-scale for determining the familiarity to a location ______65

Table 3 Setting factors and their components ______67

Table 4 Event (or dream scene) factors ______70

xvi List of scientific publications

Paper I Toward Dream Cartography: Mapping Dream Space and Content Cristina M. Iosifescu Enescu, Jacques Montangero and Lorenz Hurni

½ Published in Cartographica: The International Journal for Geographic Information and Geovisualization in December 2015, Volume 50, Issue 4, pp. 224-237. ½ Presented at the 27th International Cartographic Conference in Rio de Janeiro, Bra- zil, August 2015

Paper II Fictional volunteered geographic information in Dream Cartography Cristina M. Iosifescu Enescu and Lorenz Hurni

½ Published in The International Journal of Cartography in June 2017, Volume 3, Issue 1, pp. 76-87. ½ Presented at the 28th International Cartographic Conference in Washington DC, USA, July 2017

Paper III Place Cookies and Setting Spiders in Dream Cartography Cristina M. Iosifescu Enescu, Hans-Rudolf Bär, Matthias Beilstein and Lorenz Hurni

½ Accepted for publication in Transactions in GIS as of November 2019; for publication in 2019 in a special issue on modelling and analyzing platial representations ½ Presented as short paper Pinpointing Dream Settings onto Place Cookies at the Platial’18 Workshop in Heidelberg, Germany, September 2018

Paper IV Cartographic Tools for Mapping Dreams Cristina M. Iosifescu Enescu and Lorenz Hurni

½ Published in the Proceedings of the 29th International Cartographic Conference in Tokyo, Japan, July 2019 ½ Presented at the 29th International Cartographic Conference in Tokyo, Japan, July 2019

xvii

Chapter 1. Introduction

1 Introduction

Motivation Nowadays, the entire Earth, i.e. in the “real world”, is topographically mapped or at least immediately depicted through aerial or satellite imagery. However, it is possible to make maps based on fictional content (Hurni, 2015, p. 61). Dreams open a new world for cartography: mapping, analyzing and structuring of unknown spaces, discovering the individual worlds created by the human brain, with their eccentricities, with their liaison to a greater or lesser extent to the geographical world.

The cartography of dreams is a promising research topic, being of interest for both scientific communities in cartography and in dream research, and as well for lay people. Whereas individuals have always showed a vivid interest in their dreams, the area of dream research had ups and downs, often being considered unscientific. Dream theories, such as the cognitive approach to dreams, advocate the observation and analysis of the manifest content of dreams for gaining a better knowledge of the dream functions, of oneself, but also for being used in psychotherapy.

Geography and cartography scholars, used to seek in every phenomenon after its spatial dimension, are facing a challenge, when looking at dream content. Dream space can entirely reflect a geographical location, but can also represent mixed places, which combine distinct features of different locations, embedding personal experiences and thoughts, or distorted places, or even fictional, non-existing places.

A need was therefore identified to model the dream data and to construct a visualization method, which could fit the particularities of the spatial dimension in dreams. Previous work on data modeling and representation of fictional space in literary geography, performed in the project “A Literary Atlas of Europe” (Piatti, Reuschel, & Hurni, 2011), functioned as an incentive to setting up the Dream Cartography project. That interdisciplinary project, gathering literary and cartography scholars, resulted in a foundation for the cartography of fictional spaces and the advantages of this collaboration encouraged the planned interdisciplinary work for the current project, between the fields of psychology and cartography.

Problem Statement and Research Questions Dream researchers analyze the content of dreams for various purposes. Beyond the well-known application in psychotherapy, dreams are being analyzed also from an academic perspective, to answer questions such as what are the phenomenological characteristics of dreams, how the dream content is related to waking-life experiences or how dreams do affect the subsequent life of the dreamers (Schredl, 2010).

In this regard, describing and unraveling the spatial dimension as it is experienced in dreams is a step toward understanding the subjective experience of the space in general. The theory and instruments to acquire and visualize the dream space are lacking. In the following paragraphs, observed shortcomings are expressed and adequate research questions are formulated.

An important issue in dream research deals with the data sources and the types of data, which can be used for dream analysis. Sharing dreams is currently possible only by the power of narratives, written reports (dream diaries) or drawings. A dream cannot (yet) be directly recorded from the brain of the dreamer, even if there existed attempts to decode images directly from the dreaming brain (Horikawa, Tamaki, Miyawaki, & Kamitani, 2013) using non-invasive imaging methods such as the Functional Magneto-Resonance Interferometry (fMRI). Dream researchers rely on dream reports as primary data source, and additional surveys or questionnaires are used to gather specific data. Therefore, it is crucial to get the most accurate dream reports possible. Is there a way for dreamers to better recall and record a dream? Furthermore, if questionnaires cannot be applied, what are additional sources, which can be used for complementing a dream report? The following research question is formulated:

Research question 1 What are the sources of dream data and how can data acquisition be improved with respect to the spatial dimension?

The spatial dimension of dreams is currently scarcely addressed in dream research. Previous work related to modeling the dream content (Hall & Van de Castle, 1966; Strauch & Meier, 1996) resumes to classifying the settings as whether these are indoor or outdoor, familiar or unfamiliar, geographical, distorted, or unspecific. Is this is an oversimplification? Are there more elements that matter in a dream setting? How can elements describing the dream setting be acquired from a dream report? The following research question is formulated:

Research question 2 What are the components of a dream setting and how can dream settings be modelled by an appropriate data model?

Another issue is the visual representation of dream settings. In many domains, such as history or economy, the spatial dimension of the data can be rather easily inferred and set as a property of this data. In dreams this is different, because the dream space presents particularities: not

3 Introduction

only geographical locations are dreamed about, but also places that are mixed, distorted, uncertain, or entirely invented. The problem is that this dream space is difficult to be depicted on a map. Should the space of a dream not have its own base map, its own topographical map?

Research question 3 What are adequate cartographic methods for a visual representation of a dream setting?

For analysis, the dream content is usually split into different topics. For example, Hall and Van de Castle (1966) proposed topics such as dream characters, social interactions, emotions, goal achievement, etc. Each of these topics can be modeled differently, and specifically developed coding schemas are used by raters to code dreams. With this codes, statistical analysis can be performed over these dreams. However, for a qualitative analysis, the deepness of the dream experience is lost. A visual structuring and representation of dream topics, together or separately, can bring more clarity and reveal unexpected patterns.

Research question 4 How can the different elements of the dream content be visually represented?

There are cases, where the spatial component of dreams is not clearly expressed. However, the setting can always be characterized by its elements. Is it possible to use these elements for a visualization of the dream setting? Is a visualization of the dream content possible in a single profile? Being able to compare dreams, respectively dream settings, is a request for the exploratory dream analysis. Not only parametrical comparisons, enabled by data modeling, but visual comparisons could bring advantages.

Research question 5 How could a visual profile of a dream setting look like and how can a visual comparison of dreams and, specifically, of dream settings be realized?

The formulated research questions are addressed in the scientific publications, which are part of this thesis (Table 1). In the following, the developed and applied methodology is summarized and the content of the research articles is shortly outlined afterwards.

Table 1 Research questions covered by the publications RQ1 RQ2 RQ3 RQ4 RQ5 Data sources Dream set- Mapping a Visualization Dream set- and acquisi- ting model dream setting of dream ting analysis tion content Paper I ✓ ✓ ✓ Paper II ✓ ✓ Paper III ✓ ✓ ✓ ✓ Paper IV ✓ ✓ ✓

Methodology The identification of shortcomings and needs for the data acquisition, especially regarding the dream setting, requires the study of literature on dream research and meetings and e-mail ex- changes with experts in dream psychology (from University of Geneva and University of Bern, Switzerland; University of California, Santa Cruz, USA; and Central Institute of Mental Health, Mannheim, Germany). The gathered information is structured and formalized (paper I) into the recommendations for the improvement of dream reports and into a first proposed questionnaire, which focuses on the spatial dimension of dreams.

For a better understanding of how people recognize a place in their dreams, an empirical survey is developed (paper II). It features multiple choice questions about what are the elements used for place recognition, and also allows the introduction of free-text for the description of a lately dream location. This survey was completed online by a convenience sample of more than 300 people. Its results are used to put the basis of another list of more detailed questions about the dream setting.

One of the questions from this list – how familiar is the dreamer with the dream setting – is taken apart and modeled into the place cookie (paper III), after the circle of friends model from the literature in sociology and anthropology. This model serves as a proof-of-concept for the next, more detailed model.

The previously mentioned list of questions is enriched by studying literature in cartography, in literary geography, in GIS for humanities and narratology. A part of these questions (17) are put into a paper-and-pencil survey (paper III), which was applied to an academic group composed of 30 cartography and geoinformation specialists. The results of this second survey vali- date a good part of the questions, yet further discussions with the participants and with experts in psychology reveal that the components of a setting are more numerous than in the second survey. The careful study of dream reports (paper III) from a freely available online database1 and from published dream diaries (von Uslar, 2003; Adaman-Tremblay, 2015) enables the crys- tallization of 26 variables, which extensively describe a setting. All these variables are modeled on a single scale: from usual to unusual. The thematic grouping of these variables into factors (Paper III) results in a complex but compact data model for settings, the setting spider. A similar approach is used to create the event spider, a model of the dream topics for a dream scene. However, the dimensions of the event spider are directly extracted from dream psychology research on dream content. These dimensions are modeled on a similar scale: from not important to very important.

The two models for the setting are visualized as profiles (papers III and IV). These profiles contain information about the dream setting, which is not necessarily related to geographical coordinates. The similarities of these profiles to diagrams used thematic cartography inspire

1 http://dreambank.net/

5 Introduction

their use as uni- or multi-variate point symbols on the map. Furthermore, the characterization of the relation of dream space to geographical space leads to proposing different visualization methods (papers I and IV) of the dream setting (morphing, map collage). The visualization of other dream elements is inspired from the cited literature.

Finally, the developed models are applied for dream mapping and analysis. The requirements for a Web platform for mapping dreams are formalized (paper IV). These are implemented into the Web platform. A user of this platform answers the 26 questions of the setting spider model for a couple of his dreams and gives feedback on the model. The visualization of the setting spiders on a common dreamland map (paper III) allows the comparison of the dreams’ settings.

Overview of the Research Articles Paper I – Toward Dream Cartography: Mapping Dream Space and Content The first research paper offers a definition of the introduced research area of dream cartography and proposes a workflow on mapping dreams. It follows the processing chain established in Geographical Information Systems (GIS): data acquisition, modeling, analysis and visualization. However, there are some issues with the input data: the dreams cannot be directly recorded at the time of the dreaming. These issues are the focus in this paper, and the proposed solution addresses the possible enhancement of data acquisition in form of dream reports and of answers to specially developed questionnaires. Moreover, some visualization examples are given and explained. Two dream settings, which are a mix between two different locations, are represented in a map or in a 3D model, respectively. Visualization of two other elements of the dream content, dream characters and social interactions, are also exemplified in this paper.

Paper II – Fictional Volunteered Geographic Information in Dream Cartography The extraction of geographical information from a dream report is not trivial. Due to the characteristics of dreaming to deform spatial relations or to combine elements from different biographical memories and different sources into one dream (Fosse, Fosse, Hobson, & Stickgold, 2003), an automatic retrieval of geographical information can hardly be applied on dream reports. Therefore, a new type of Volunteered Geographical Information (VGI) is proposed in this paper: the fictional VGI (F-VGI). The social aspect of dreams is exploited, arguing that not only the dreamer, but also a third party may contribute to the fictional geography. Externally available additional data sources are discussed, the scale of the space in dreams is mentioned, and examples of dream settings from a survey on dream locations are given.

Paper III – Place Cookies and Setting Spiders in Dream Cartography This paper addresses the complex issue of modeling the dream setting. Beginning with a straightforward model, the place cookie, it continues to a complex dissection of the setting, revealing its components. This latter model is used to create a visual profile of the dream setting, named setting spider. Moreover, a similar model is built for a dream scene, resulting in the so- called event spider. The proposed profiles enable the comparison of different dream settings,

or of different dreams, respectively. This is proven with the help of a personal dreamland map, on which the setting spiders of eight dreams are depicted as multivariate point symbols. It is argued that the proposed models can be adopted also for real-life settings and events.

Paper IV – Cartographic Tools for Mapping Dreams The methods proposed in the previous papers are implemented in a Web Platform for Mapping Dreams, which was implemented as a functional prototype. The requirements for a dream mapping platform are formalized, and the current and other possible implementations are discussed. In a case study of a dream it is shown, based on data acquired through a Web form, how the spider profiles are generated, how a dream map can be created and modified, and how other dream elements can be visualized.

Relevance for Science and Society The results of this project can bring benefits to several areas of scientific research: cartography and GIS, psychology and cognitive sciences, and literary studies. For cartography, mapping dreams extends the mapping of “different geographies” to the realm of dreams. Data modelling requires in the first-place consolidated knowledge of the data. This thesis points out that data acquisition can be improved by adding other data sources and by actively asking the producer of data for structured data, in this case by applying questionnaires. Regarding a possible base map for dreams, this project dares to cut apart and recombine the real geography, for depicting mixed locations in dreams. This prepares map readers to new, unconventional ways of handling maps. The proposed model of a setting is of interest also for GIS, because it formalizes the subjectivity of perceived places. Reconnecting the setting model to the original dream report can reveal individual differences in the human cognition of space and could be used in location- based services to create adaptable, subjective instructions for navigation.

For dream psychology, this project offers the means to explore the spatial dimension in dreams. It proposes models for both dream content in general and for the dream settings. The coding of dreams on these models is done by the dreamers themselves at the point of data acquisition by answering straightforward questions. This way, it bypasses a possible rater bias. The models allow the comparison of dream settings, respectively of dreams (or dream scenes). Using them for a series of dreams can reveal patterns and can be used for an exploratory analysis. Memories of biographical events can also be a case of study, where the models may be applied. In this case, the questionnaires would not refer to a dream that just happened, but to a memory of an event. Moreover, applying the models to specific personal life-important events can be relevant to the study of autobiographical, episodic memory in psychology. The developed models can be used also in literary studies, to enhance the understanding of a literary setting and allow comparisons between the writing styles of different authors.

The data models can help organizing and offer a structure to a personal dream diary. The proposed fictional VGI and the Web platform for mapping dreams offer to a broader range of users

7 Introduction

an interesting tool for visualizing and organizing their dreams. This can contribute to an increasing engagement of people with their dreams and especially with the spatial dimension of dreams. A side-effect could be a better awareness of space in general, and acknowledging personal experiences as valuable.

Structure of the Thesis This cumulative dissertation is structured into five chapters. Chapter 1 covers introductory aspects such as the motivation of the dream cartography project, the problem statement and the resulted research questions. The developed and used methodology is shortly described and the scientific and social relevance of the conducted research is estimated. Chapter 2 provides background information about relevant research areas on fictional, human space, and on dream analysis.

Chapters 3 and 4 form the main part of the thesis and comprise the four research papers. Chapter 3, Dream Cartography Concepts, first deals with qualitative analysis: data modeling, data acquisition and visualization examples of aspects appearing in a single dream (papers I and II). Paper III extends the data modeling, allowing a quantitative analysis of a dream series, and proposes an appropriate cartographic visualization method. Chapter 4 illustrates how the cartographic tools for mapping dreams, developed in the previous chapter, can be applied in a Web portal (paper IV).

In Chapter 5, the conclusion is drawn by reconsidering the formulated research questions and summarizing the results. Finally, an outlook on further possible research related to Dream Car- tography is given.

Chapter 2. Background

This chapter contains excerpts from papers I, II and III.

9 Background

Different Geographies Dream Cartography can be seen as a propagation of the spatial turn in the humanities to the domain of dream research. The spatial turn (Warf & Arias, 2008) was a movement, where the spatial dimension became relevant for many disciplines other than geography and cartography. This gave birth to interdisciplinary research between geosciences and humanities, where the focus is laid on the human space, the space as it is perceived by people (see the next section). Consequently, powerful tools of information for historical, cultural or artistic phenomena are created, complementing the digital humanities.

An example of such interdisciplinary research is literary geography, which deals with the representation of space in fictional, literary works (Moretti, 1999; Piatti, 2008). In 2011, an entire issue of the Cartographic Journal was dedicated to literary geography (“Cartographies of Fic- tional Worlds,” 2011). In the project “A Literary Atlas of Europe”, the available space information is modeled into individual spatial entities according to their function: settings, zones of actions, projected spaces, routes and markers (Piatti, Bär, Reuschel, Hurni, & Cartwright, 2009; Reuschel, Piatti, & Hurni, 2013). Moreover, “places with uncertain, transformed or hardly locatable spatial objects necessitate additional attributes […] in order to reflect the fictional world” (Reuschel et al., 2013, pp. 142–143). Beside geometry, attributes are proposed such as the uncertainty degree (precise or zonal) and the relation to the actual geospace (imported, transformed, invented or imagined). For the cartographic representation of single spatial objects in the “Literary Atlas of Europe” are proposed special visualization methods such as fuzzy shapes or concentric arcs for uncertain places. Shifted spaces are brought in connection to their true location using directed Bezier curves (Reuschel, 2012; Reuschel & Hurni, 2011). For historical novels, both the current geography and historical maps are displayed. The proposed attributes and visualization methods can be adapted for the dream space.

Literary geography can be classified as a “different geography”, where “maps are used to trans- mit information about non-real places using the same methodology as applied to mapping real places” (Kriz, Cartwright, & Hurni, 2010, p. 1). The prime example of a different geography is the London Underground Map from Beck, 1931, that modified the real geography to help navigating in the real space. The book “Mapping Different Geographies” (Kriz, Cartwright, &

Hurni, 2010) offers an overview on this subject, mentioning also imagined places and dreams. Namely, the case of a real place first learned about in school, then enhanced by the study of a visual atlas, and finally experienced in a dream, can already create a strong connection, without a physical presence to this place (Vaughan, 2011).

The different geographies are part of a broader, umbrella-term of “art and cartography”. This deals with subjects such as literary geography, cinematic cartography, personal geographies or humorous and caricature maps (e.g. for political propaganda) by means of artistic interpretations and representations of space (e.g. Harmon & Harmon, 2004; Baynton-Williams, 2015; Lewis-Jones, 2018; DeGraff & Harmon, 2015; DeGraff & Jameson, 2017). Here, cartography is set as an “interface between art and place” and the purpose is to examine different narrative structures, to understand places as they appear in artworks and to explore personal and emotional relationships to places (Ribeiro & Caquard, 2018).

Human Space The debate on the meaning and understanding of “space” versus “place” and the creation of place as the space directly experienced by people is addressed in human geography by Tuan (1977) and later by Cresswell (2014). Space becomes place through the interaction of a person with it, but also through a story: “Isn’t it strange how this castle changes as soon as one imagines that Hamlet lived here?” (Tuan, 1977, p. 4). Dennerlein (2009) analyzes how the space/ place/ setting is narrated and what elements are needed for a place to be defined in a literary text. It is known that human cognition of space is different from the Euclidian geometry (Frank, Mark, & Raubal, 2013). How people perceive space and time and, subsequently, how they refer to and reason about these, is the subject of naive geography, defined as “the body of knowledge that people have about the surrounding geographic world” (Egenhofer & Mark, 1995, p. 4).

Consequently, spatial (and temporal) relationships can be absolute, relative and intrinsic (Ten- brink & Kuhn, 2011). Lefebvre advocates that space must be understood not only as a concrete, material object but also a lived, subjective one (Lefebvre, 1974/1991). Tilley (1994, p. 8) summarizes some characteristics of abstract, scientific space compared to the humanized, meaningful space (Table 2), concluding that, other than previously assumed, the abstract space is the irrational, and the human space is the rational one. Therefore, the human space is a medium, which can influence the action, whereas the abstract space is just a container. The abstract space is a “simple surface for action, lacking depth”, which makes it interchangeable with any other space. Its impact on the society is decoupled of time and culture.

Moreover, the human space is centered on the person – compare also to Bollnow (1963). The Euclidian space is homogeneous and with coordinates centered at an arbitrary point, whereas the experienced space is heterogeneous and has its implicit coordinates in the axes of the person. Bollnow also remarks that places are qualitatively different for people, and, based on these

11 Background

qualities, a personal classification of places comes into being, which has no counterpart in the abstract, mathematical space.

Table 2 Differences between abstract and human space from Tilley (Tilley, 1994, p. 8) ABSTRACT SPACE HUMAN SPACE Container Medium Decentrated Centred Geometry Context Surfaces Densities Universal Specific Objective Subjective Substantial Relational Totalized Detotalized External Internal System Strategy Neutral Empowered Coherence Contradiction Atemporal Temporal

The human cognition of space appears also in the context of globalization. Prototypical places are created and sustained by cultural phenomenon, such as food (e.g. Italian or Chinese restau- rants), or by global vision or even interior decoration of multinational companies, being totally dislocated from the respective geography. In this regard, in dreams, literary science and storytelling, a different concept of place is used, the “setting”. Steele (1981, p. 11) defines the setting as “a person’s immediate surroundings, including both physical and social elements”.

Settings are characterized by different elements. For Foucault (1967/2008) the space can be private or public, cultural or useful. Lefebvre (1974/1991) argues that the (social) space is a social product, and that space embodies social relationships. Therefore, important elements characterizing a setting are the culture, or the spoken language, but also a possible contradiction between the place and the circumstances. In a globalized world, not only the culture can make a difference, but also the wealth, the income: e.g., high income classes of different cultures may possibly have more related standards and views than high- and low-income classes of the same culture (Rosling Rönnlund, 2016). Finally, Montello (1993) identifies different scales of experiencing space: when space is smaller than the human body, such as distant landmarks or maps; or when the space is larger than the human body and therefore, the space surrounds the body.

Dream Research Dreaming is defined as “a state of consciousness characterized by internally-generated sensory, cognitive and emotional experiences” (Desseilles, Dang-Vu, Sterpenich, & Schwartz, 2011) which occur usually during sleep, but also on so-called states of day-dreaming.

The Cognitive Approach studies the form or content of dreams and explains these “with references to cognitive processes involved in waking and sleep mentation” (Montangero, 2012a). Some cognitive studies are quantitative content analyses on large samples of dreams (Domhoff, 1996; Strauch & Meier, 1996); for example they specify the percentage of the studied dreams which take place indoor versus outdoor. Other cognitive studies focus more on dream processes and deal with qualitative aspects of dream content, such as the development of children’s dreams (Foulkes, 1999), the narrative organization of the dream experience (Cipolli & Poli, 1992; Montangero, 2012b) or the nature of memory sources of dreams (Baylor & Cavallero, 2001). According to researchers using a cognitive approach (Domhoff, 1996), dreams reveal conceptions and current concerns of a person.

In dream research, the dream setting is so far rather scarcely treated. The established data models resume to classifying the dream setting into familiar versus unfamiliar or indoor versus outdoor. Table 3 shows the coding of the element setting from the dream content in the Hall and Van de Castle System (Hall & Van de Castle, 1966; Domhoff, 1996).

Table 3 Coding dream settings as proposed by the Hall/Van de Castle coding system2

LOCATION FAMILIARITY D Distorted I Indoor F Familiar O Outdoor G Geographical A Ambiguous U Unfamiliar NS No setting Q Questionable

These coding letters are combined, and some examples are shown in the following3:

2 From https://dreams.ucsc.edu/Coding/ 3 examples taken from https://dreams.ucsc.edu/Coding/settings.html

13 Background

IF “I was in my room getting dressed”

ID “It looked like my history classroom except the desks were of kindergarten size.”

IG “I looked out the hotel window and saw New York City below.”

IQ “I was in a store buying a pair of shoes.”

OG “We were swimming as some Hawaiian beach.”

OQ “The football field we were playing on was muddy”

AF “The view of the Eiffel Tower was magnificent.”

In the familiarity column, there is a precedence order, the distorted being mentioned if it occurs, in the detriment of familiar, geographical or unfamiliar. Therefore, the code D from distorted takes precedence over any other code from the familiarity column. Geographical G is used for the places indicated by their geographical names. However, if the dreamer indicates, that this place is familiar, the F code is being used, because the familiarity to the dreamer is more important. Unfamiliar settings can be coded as familiar in case of famous settings (see example with Eiffel Tower) or as Geographical, for the case they are named with a geographical name. Questionable is used where it cannot be determined from the dream report if the setting is familiar or not. The last example is coded as ambiguous with respect to the indoor or outdoor classification, because the Eiffel Tower can be contemplated both from an outdoor location or from an indoor location, through a window.

In the questionnaire on dream content proposed by Bernstein and Roberts (1995) the indoor- outdoor paradigm is addressed with “restrictive and confining” versus “spacious and vast”. The authors assess also if the setting changes or remains constant during a dream.

Traditionally, the dream content is evaluated by trained raters, sometimes also by asking the dreamers to complement the findings. However, the performance of raters in the case of dream content is rather poor. The agreement of two independent raters over the content of the same dream, the interrater reliability ranges from 44% to 87% in different studies (Hall & Van de Castle, 1966; Sandler, Kramer, Fishbein, & Trinder, 1969; Schredl, Burchert, & Gabatin, 2004). This depends on the raters’ experience, on the number of analyzed dreams of the same dreamer, on the used scale and on the studied element (emotions, settings, dream characters, etc.).

Regarding the incorporation of familiar settings in dreams, Strauch and Meier (1996) classify the settings as familiar, distorted, unknown or non-specific. The analysis of 500 dreams with 642 different settings reveals the rate of familiar settings appearing in dreams to be of about 26% (see Figure 1).

non-specific locales 19% familiar settings 26%

distorted surroundings 11% unknown scenery 44%

Figure 1 Dream settings, subdivided according to their familiarity; from (Strauch & Meier, 1996, p. 108)

The Web as a Communication Medium The Web had a great influence in changing maps from static entities into dynamic, interactive products (Peterson, 2008). Not only is the interactive access to customized map content possible (e.g. Iosifescu Enescu, Iosifescu Enescu, & Hurni, 2013), but also the process of mapping is now open. In this regard, Goodchild (2007, p. 211) designates Volunteered Geographic Infor- mation (VGI) as the harnessing of Web tools “to create, assemble, and disseminate geographic information provided voluntarily by individuals”.

Many people show a strong interest in their dreams and this is observable on the Web. Openly available dream reports are found not only in published books and dream diaries (e.g., von Uslar, 2003; Adaman-Tremblay, 2015), but also in dream forums (klartraum.de, dreamviews.com, dreamscloud.com) or in scientific online dream databases (dreambank.net). Moreover, dreams are openly discussed on social media (e.g. the group of the International Association for the Study of Dreams on Facebook). This happens although dreams are very personal experiences and they reveal much of the current concerns and activities of the dreamer, as the continuity hypothesis of dreaming and its experimental basis sustain (Strauch & Meier, 1996; Domhoff, 1996; Erlacher & Schredl, 2004).

Contribution to Research This doctoral thesis contributes to research by connecting these previously disconnected topics, for which a brief background was provided above. Therefore, mapping the dream world is mapping a Different Geography, and, consequently, deals with design of information related to location. The setting experienced in dreams, being centered on the person, has the properties of

15 Background

human space, rather than of abstract space. The dream research deals so far scarcely with understanding the composition of dream settings, and the proposed solution is to approach dreams from the perspective of geography and cartography. Regarding the Web as a communication medium, VGI and dreams in social media are the triggers for the theory, methods and technology that are developed in this work, for creating personalized maps for dreams.

The following chapters present in detail the research conducted in the frame of the dream cartography project. The spatial dimension of dreams is empowered hereby, just as every story takes place somewhere, even dreams.

16 Paper I

Chapter 3. Dream Cartography Concepts

17 Dream Cartography Concepts

Paper I. Toward Dream Cartography: Mapping Dream Space and Content This is the Author’s Original Manuscript (AOM) of an article published by Springer in Carto- graphica, The International Journal for Geographic Information and Geovisualization in De- cember 2015

Cristina M. Iosifescu Enescu*, Jacques Montangero**, Lorenz Hurni*

*Institute of Cartography and Geoinformation, ETH Zurich, Switzerland **Department of Psychology, University of Geneva, Switzerland

Abstract. The objective of dream cartography is to map dream content, that is, to develop new visualization methods and adapt existing ones as graphical means for an explorative dream analysis. We aim at modelling, documenting, reuniting, and graphically depicting the dream space and other relevant psychological aspects of dreams in a meaningful but abstracted way. In this article we describe the proposed workflow for mapping dreams, focusing on data acquisition, and illustrate it with some proof-of-concept visualization examples. The dream space is fundamentally different from the real space, especially because it is not possible to consciously visit the place of the dream again. One way to address this setting issue is through data acquisition, and here we propose means for getting a more accurate description of the dream space and of other important dream elements.

Keywords: Dream Cartography, Dream Space, Condensed Places, Data Modeling, Cartographic Information Systems 1 Introduction Cartography describes places and spatially distributed phenomena graphically, using symbolization. Psychology is based on the description of people’s behavior and mental content. Dream Cartography brings cartography and psychology together, in order to get new insights, through cartography, on the subject dream. Dream is a topic that concerned generations of psychologists

18 Paper I

and is still a broadly active research area nowadays. How do dream elements (places, social interactions, emotions, goals, etc.) look like, how do they differ from the real life? Dream space, especially, presents a cartographic challenge because of its particularities: not only geographical locations are dreamed about, but also places that are mixed (which reunite distinct features of different places), condensed (as mixed, but embedding also different personal experiences and thoughts), distorted (spatial proportions are different than in reality), uncertain or totally invented. Moreover, social interactions, dream storyline and emotions, to name just a few, are important psychological aspects, which contribute to the complexity of the inner-world to be mapped.

In this context, Dream Cartography proposes the development of new visualization methods and adaptation of existing ones for dreams in order to create maps for mixed, condensed and distorted places, map uncertain geodata, visualize relative position respectively graphically represent time flow and character occurrences, social relationships or valence / emotions arising in dreams. The goal is to integrate all these visualization elements into an interactive Web platform, so that each dream presents itself to the map viewer in its actual multidimensionality. Our project targets especially dream researchers and psychology scholars, as well laymen interested in their dreams.

The cartography of dreams opens up new dimensions of research, in the domain of classical cartography, GIS and of psychology. The first scientific relevance is in cartography, since this means exploring new, undiscovered territories, creating very specific, individual maps and proposing novel symbolization. The insights that we gain from the visualization of the spatial dimension of dreams may be used to visually represent other data or descriptions of places, as the surrounding world is interpreted by human cognition. The latter might be used in location-based services in order to improve for example verbal instructions for navigating to unknown places.

From the point of view of psychology, we add our cartographic contribution to the cognitive theory of dreams, where the manifest content of dreams has significance in itself. Accordingly, we can provide a user-friendly, graphical tool for exploring the dream storyline and landscape, allowing end-user to adapt the visualization in order to reflect the importance of the occurring elements in the dream. Dream Cartography c serve as an important analysis tool for psychologists and psychiatrists who may explore individual dreams and dream series of their patients for discovering patterns in the dreams, such as reoccurring places, emotions, type of characters or of social interactions.

In this paper, we first state the problem of space creation in natural language, then give some compact background information about dream research and continue with the embedding of dream cartography into state-of-the-art cartographic research. We finally sketch the workflow of producing dream maps based on some given examples and outline the most challenging aspects of this process that require further investigation.

19 Dream Cartography Concepts

2 Setting as a Basic Construct in Dream Cartography The debate on the meaning and understanding of space versus place and the creation of place as the space directly experienced by people is addressed in human geography by Tuan (1977) and later by Cresswell (2013). Furthermore, Dennerlein (2009) analyzes how the space/place/setting is narrated and what elements are needed for a place to be defined in a literary text. It is known that human cognition of space is different from Euclidian geometry (Frank, Mark, & Raubal, 2013). How people perceive space and time and, subsequently, how they refer to these, is the subject of naive geography, defined as “the body of knowledge that people have about the surrounding geographic world” (Egenhofer & Mark, 1995, p. 4). There- fore, spatial (and temporal, for that matter) relationships can be absolute, relative and intrinsic (Tenbrink & Kuhn, 2011). Concerning the topic of dreams, Forshage (2007) addresses the issue of “how space is constructed” in dreams, using identifiable criteria from “natural phenomena (flora, fauna, geomorphology, meteorology, etc.)”. Related to his novel, a Swedish non-scientific community makes interesting observations on spatio-genesis in dreams (“The landscape of our dreams - foundations of dream geography,” 2011) and names superordinate aspects of the spatial phenomenology such as the geometry and structure of the dreamscape, the spatial relations, the means of recognition and orientation (e.g. what signs are used for recognizing a place: signposts, spoken language, infrastructural characteristics, etc.), active integrative learning from dream abilities, experiences and solutions (e.g. “solving practical problems by rear- ranging spatial relations”). Indeed, there are “different geographies” and, starting by mentioning the London Underground Map from Beck, 1931, that modified the real geography to help navigation into the real space, the book “Mapping Different Geographies” (Kriz, Cartwright, & Hurni, 2010) offers an overview of these, which, being removed from conventional maps, are nevertheless powerful tools of information for historical, cultural and artistic phenomena. An example of a “different geography” is the “literary geography”, which we discuss closer in the section 4.1.

The term of uncertain geodata was first introduced Geoinformation Science in order to take data quality into consideration when visualizing geodata (MacEachren, 1992). For uncertain representations, an extended set of graphical variables are proposed: color saturation, crispness of symbols, transparency and resolution of raster geodata (MacEachren et al., 2005). The visualization of uncertain geodata is an important topic for instance in the emerging field of literary geography, where uncertainty is related more to the vagueness and subjectivity of data (Re- uschel & Hurni, 2011) than to the quality of input data.

In comparison, the work that has been done in the dream research community related to settings, on which the dreams take place, is mainly related to classifying these into indoor/outdoor or with respect to their familiarity to the dreamer (e.g. using the Hall and Van de Castle [1966] dream coding system; Domhoff, 1999). It has also been noted that a specificity of dream space, beside condensations of elements of different places in one dream location, is the selection of a few elements of a totality or the representation of one part of the setting only (Montangero,

20 Paper I

1999). This can make the same setting look different to different persons, which for example in computer gaming is called phasing4. 3 Dream Research: a Brief Overview Dreaming is defined as “a state of consciousness characterized by internally-generated sensory, cognitive and emotional experiences” (Desseilles, Dang-Vu, Sterpenich, & Schwartz, 2011) which occur usually during sleep, but also on so-called states of day-dreaming.

The most prominent approaches to the study of dreams nowadays are the Psychoanalytic Dream Theory and the Cognitive Approach to Dreams. The Psychoanalytic Dream Theory focuses on dream interpretation in the sense that the dreams disguise wishes but also reveal the way persons process conflict and emotion, cope with anxiety, their defense mechanisms or their creative potential (Freud, 1942; Hau, 2006). It is Freud who first proposed the term “condensation” in dreams, which he describes as the bringing together of different aspects of the dreamer’s experience into a single dream image (Freud, 1942). The Cognitive Approach studies the form or content of dreams and explains these “with references to cognitive processes involved in waking and sleep mentation” (Montangero, 2012a). Some cognitive studies are quantitative content analyses on large samples of dreams (Domhoff, 1996; Strauch & Meier, 1996); for example, they specify the percentage of the studied dreams which take place indoor versus outdoor.

Other cognitive studies are more interested in dream processes and deal with qualitative aspects of dream content, such as the development of children’s dreams (Foulkes, 1999), the narrative organization of the dream experience (Cipolli & Poli, 1992; Montangero, 2012) or the nature of memory sources of dreams (Baylor & Cavallero, 2001). According to researchers using a cognitive approach (Domhoff, 1996) dreams reveal conceptions and current concerns of a person.

Apart from the dream research focusing on dream content, which we specifically consider for the current project, we mention the interest of neuropsychology in brain activation during sleep, respectively during dreaming (Nir & Tononi, 2010). Neurophysiological (Electroencephalo- gram) and neuroimaging (Functional Magneto-Resonance Imagery) methods generate maps of brain activation. These visualize areas of the human brain in the attempt of understanding the dreaming process from the physiological point of view, which is fundamentally different to mapping the dream content as we propose in the current project. 4 Integrating Dream Cartography into Current Cartographic Re- search The cartographic aspects of mapping dream contents cover the modeling of input data for an automatic generation of maps, the visualization of dream space with its specificities in the form of maps, the developing of comprehensive visualizations schemas (e.g. diagrams or network-

4 http://www.wowwiki.com/Phasing, accessed on 15.11.2014

21 Dream Cartography Concepts

like representations) for other dream variables and finally the integration into a cartographic information system, respectively an interactive, Multimedia Atlas Information System (Hurni, 2008). In the following, we restrict our observations to current research on modeling and visualization of uncertain geodata as addressed by the literary geography and to interactive Web platforms and Atlas Information Systems, which represents the way in which we intend to com- municate dream maps to our target audience. 4.1 Literary Geography Regarding the modeling and visualization of uncertain geodata, we mention the research work that has been done in the emerging field of the literary geography, which deals with the representation of space in fiction. In order to better represent the literary space, our colleagues from the project “A Literary Atlas of Europe” modeled the available space information into individual spatial entities according to their function: settings, zones of actions, projected spaces (which are located in a character’s mind, without the character being physically there), routes and markers (Piatti, Bär, Reuschel, Hurni, & Cartwright, 2009; Reuschel, Piatti, & Hurni, 2013). Moreover, “settings and projected places with uncertain, transformed or hardly locatable spatial objects necessitate additional attributes and composed geometries respectively in order to reflect the fictional world” (Reuschel et al., 2013, pp. 142–143). Therefore, and also for the depiction of the setting’s importance to the literary work, Reuschel et al. (2013) propose, beside the geometry, attributes such as the uncertainty degree (precise or zonal) and the relation to the actual geospace (imported, transformed – shifted or synthesized places, invented or imagined). These literary settings attributes can be adapted for classifying the importance of space in dreams as well.

For the cartographical representation of single spatial objects, the “Literary Atlas of Europe” uses symbols and labels to indicate the extent and the preciseness of the settings and color codes for the relationship between literary space and real space. Reuschel and Hurni (Reuschel, 2012; Reuschel & Hurni, 2011) propose visualization methods such as fuzzy shapes, animated or texture (radial) visualization for settings; diffuse Gaussian boundaries, fading in and out for routes and concentric circular arcs for uncertain routes’ start- or endpoints; shifted spaces are brought in connection to their true location using directed Bezier curves. Moreover, for some special cases both the current geography and the historical map from that time are displayed, which results in a better understanding of the novel’s spatial dimension.

Although the “Literary Atlas of Europe” is related to our proposed work regarding the modeling of spatial information and visualization of uncertain geodata, it deals only scarcely with mixed, condensed or distorted places. Along with dream setting, in Dream Cartography we take into consideration and visualize additional information such as the previously mentioned relevant psychological aspects of dreams, which are crucial in expressing the multidimensionality of a dream. Moreover, when the literary geography deals with fictional places (e.g. J.R.R. Tolkien), the text of analysis has a different scale, different narrative properties and places have an important feature, which dreams do not always present: stability.

22 Paper I

4.2 Interactive Atlases and the Web Multimedia Atlas Information Systems are “systematic, targeted collections of spatially related knowledge in electronic form, allowing a user-oriented communication for information and decision-making purposes” (Hurni, 2008). These do not only combine high cartographic quality and user-friendliness with special interactive functions for geographic and thematic navigation, querying and analysis, but also enrich the user experience with multimedia information such as images, video and audio documents, animations, graphics and diagrams.

Furthermore, the Web as a communication medium had a great influence in changing maps from static entities into dynamic, interactive products (Peterson, 2008). A way to integrate the power of atlas information systems with the Web is by using Web standards such as Hyper Text Markup Language (HTML), eXtensible Markup Language (XML), Scalable Vector Graphics (SVG) and Java Script/ECMA Script, whereas the cartographic representation on Web can, for instance, use the Web Map Service (WMS) protocol along with Styled Layer Descriptor (SLD) (I. Iosifescu Enescu, 2011).

GeoVITe is an example of an interactive Web platform for providing researchers with an easy- to-use online access for visualization and download of geodata (C. M. Iosifescu Enescu, Iosifescu Enescu, Jenny, & Hurni, 2011). This platform was extended for other projects such as SwissExperiment/OSPER (Open Support Platform for Environmental Research), which en- hances the visualization, among others, with time navigation and interactive, customizable symbolization of thematic data (C. M. Iosifescu Enescu, Iosifescu Enescu, & Hurni, 2013).

The different graphical elements proposed by the Dream Cartography can be integrated into an interactive Web platform such as the one for GeoVITe, along with the actual feed of new dream reports into the database and so enhance user participation and let the explorative analysis of the dream data be most appealing. After this brief review of literature related to our work, we get into the more specific part of this paper, with the proposed workflow detailed in section 5 and some proof-of-concept visualization examples in section 6. 5 Mapping Dreams: Workflow The workflow for Dream Cartography covers the following topics: data acquisition, data modeling, setting visualization and visualization of other psychologically relevant dream aspects, and, as a further objective, comprises the visualization of dream series of a person. Advanced data acquisition, a part of data modeling and the dream visualization make use of a specifically created and tailored Web GIS application, designed based on our previous experience with Web portals with customizable content and visualization (C. M. Iosifescu Enescu et al., 2013), as previously discussed in section 4.2. 5.1 Data Acquisition In the process of mapping dreams, the question of how to collect dream data becomes a key issue. The dream research community uses mostly self-written dream reports (Nir & Tononi,

23 Dream Cartography Concepts

2010). The other possibility, of getting people in a sleep laboratory and wake them up during the night in order to get a more “unbiased” dream story, is beyond the means of this work. Therefore, our options are to work with available dream reports or to gather dream reports from willing participants. In the first option, dream databases (Domhoff & Schneider, 2013) or dreaming groups organized online can provide a good starting source, and also valuable are published books which contain dream logs (von Uslar, 2003). However, the latter option brings additional value in our case, for reasons of dream recording preparation (see Table 1), submission of fresh information, follow-up questions for details (see Table 2), but also for reasons of interactivity of the dreamer with project’s Web platform and therefore possible contribution in the cartographic representations, which we address in section 6. In the following we feature this targeted data acquisition method.

Every dream takes place in a setting, but sometimes the dream location is not mentioned in the dream report (see also section 2). Therefore, beside the simple submission of a dream report, we prepared a questionnaire (see Table 2) that reminds the participants about recording the dream location. Moreover, for the targeted acquisition of dream reports, the participants are given guiding instructions on how to keep a dream log, in order to get most of the dream content. Table 1 summarizes these instructions.

Table 1 Instructions for keeping a dream log

Metadata Instruction for the dreamer Preparation of the Prepare a computer file or a paper notebook if preferred, in order to recording means write down the dream reports. Prepare also an audio recording means (mobile phone or audio recorder). Choosing an un- Decide at which date you will attempt to recall and report a dream: a stressful morning morning without time pressure, e.g., during weekends or holidays. Self-preparation in The evening before that date, when going to bed, give yourself the the evening before instruction: When I wake up tomorrow morning, I won’t move and I will immediately ask myself the question: What was on my mind just before I woke up? Dream rehearsal Upon awakening, if a dream scene is recalled, rehearse it (visualize and feel the other impressions of that dream scene again). This and the following instruction are to be completed before moving, getting up and switching the light on. More than one Then try to recapture what was happening before, or possibly after the dream scene first scene recalled and rehearse each of these other scenes. Eventu- ally “relive” all the remembered scenes of the dream, following the order of succession of the dream events, which is usually very clearly memorized.

24 Paper I

Oral recording Only after that recapitulation, switch the light on, switch the recorder on and record an oral description of the dream. You know that questions will be asked to you about the setting of the dream, therefore pay attention to its features (or to the fact that it was not clearly represented in the dream). Written report Later on, in the morning, write down the transcription of your oral report, as close as possible to its oral version. Questionnaire on Fill in the given questionnaire for each dream scene (see Table 2). dream setting

These instructions describe an ideal case, when the dreamer takes the time to focus on the dream activity and closely report a dream. Certainly, there are also spontaneous or salient dream re- calls, where nevertheless some of these steps might help to get a more accurate report. Salience refers in this context to the personal importance, but also to the visual richness of a dream (Kerr, 1993).

In order to specifically focus the attention of the awaken dreamer on the dream location and on the other psychologically relevant aspects for each dream scene, we propose the following questionnaire (Table 2), to be completed after the dream log.

Table 2 Questionnaire for a dream scene

Questionnaire for a dream scene Please write your thoughts about the remembered dream scene, including elements from the real life which you think might be connected to it. Then answer the following questions, if they apply. It is always possible to answer “I don’t know”, “I don’t remember”, or “It was not clear in the dream”. Is the dream location: visualized / known without visualization / completely indeterminate? Is the location identifiable or not? Is the location familiar or not? Is the geographical situation specified? If yes, what is it? If visualized: What was clearly distinct / vague / fuzzy in the setting? If clearly visualized: is there a condensation of features of two or more places? If yes, which ones? Does the location correspond: to your current waking location/ to a past location / to a visited location/ to a place where a relative or friend lives? Which of the following categories does it enter: built up / countryside / hills / mountains / seaside / exotic / fantasy / science fiction? Does it enter a subcategory? (e.g. street, in front of a shop; meadow, path; etc.) Could you orient yourself easy in the setting / did you have difficulties with the spatial orientation?

25 Dream Cartography Concepts

What memories (if any) come to your mind about such a location? Describe the memory and specify your ideas or emotions linked to that memory. If the location is indeterminate, can you think of elements that could nevertheless materialize the dream setting? (e.g., spoken language, signposts, infrastructure, natural phenomena) Please check that your dream report describes everything you visualized or you had the feeling to know during the dream experience about the setting of the dream scene. You must not be surprised if you were aware of only a few aspects of the setting. Was the dream scene emotionally neutral / pleasant / unpleasant? (Specify the emotion) Were you trying to achieve / avoid something in the dream? Did you have the feeling to succeed / fail in something? If yes, in what? What senses did you use in the dream: vision / hearing/ touch / smell? Feel free to add comments, if you wish.

There are certainly other important dream elements (see also Table 3), which are not covered by this questionnaire. This has two reasons: first, the questionnaire may not be too long, otherwise people might feel inhibited to fill it; second, important elements in a dream scene (e.g. the social encounters) are usually most prominent in the dream report itself (as a narration) as well as in the awake thoughts about the dream (see the first question in Table 2). 5.2 Data Modeling The dream reports are in form of narrated natural language. To start mapping the dream content, we first need to get the necessary information from the input data. Our approach is data modeling. Therefore, similar to Reuschel, Piatti and Hurni (2013, p. 148), we create different classes of space types for the general location information in dreams, such as geographically related or not, condensed or mixed places, etc.

Space is just one dream element and other elements such as social interactions, goal pursuit, emotions or time flow, deserve as well closer attention. Table 3 summarizes important criteria along with guiding questions for their demarcation in dreams.

Table 3 Criteria proposed for dream visualization and considerations made toward their modeling into parameters

Proposed criteria for dream visualization Dream place / space / location Is the setting of the dream scene identifiable? How clear is the location represented? Is it a complete or a partial view of the setting? A smaller or a larger area? Is it familiar to the dreamer? What type of landscape does it represent? Is there a condensation of two or more places? A comparison to a “known”5 place? How significant is the setting in the dream scene?

5 E.g. „It was like in the Lord of the Rings”

26 Paper I

Social Interactions Are there any social interactions in the dream scene? Which is the nature of the interactions? Are they reciprocal? How do the social interactions change across the dream? Are these positive, neutral, or negative? Actors Are there any other actors in the dream scene? Are they familiar to the dreamer? Masculine or feminine? Are they dominant, caring, helpful? Do they appear continuous in the dream scene? Where are they in the real life (if applicable)? Goals Does the dream scene contain any goal directed behavior? Is it a step done towards or away from goal achievement? Are there any obstacles encountered? How does the dreamer feel about the goal? Does the perceived goal change throughout the dream? Perceiving of self Takes the dreamer active or passive part to the dream? What is the emotional state of the dreamer? Where is the dreamer in the real life? With which senses does the dreamer perceive the dream? Emotions Is the overall emotion related to the dream scene positive or negative? How strong is this feeling? Time flow Does the time flow continuously? What is the time frame of the dream scene? And of the whole dream? How do the dream scenes flow?

Based on these questions we prepare a detailed modeling schema of the dream elements into parameters, by adapting the literary space model as proposed by Reuschel et. al (2013) and the coding system developed by Hall and Van de Castle for the quantitative analysis of dreams (Hall & Van de Castle, 1966; Domhoff, 1996). These are supplemented with specific schemas for additional elements and features, as needed. The resulting parameters are saved for each dream and dream scene into the project’s database in order to be retrieved for visualization. If a dream element cannot be automatically modeled based on the dream report, questionnaire and modeling schema, then, in the ideal case of targeted, interactive dream acquisition, the dreamer might be requested to select a category or concrete specify or add an element. 5.3 Visualization Dream Cartography proposes different visualization methods to be chosen from and combined into a dream illustration. For special dream space we produce location maps with special dif- ferentiation and creation method for mixed, condensed and distorted places. For vague or fuzzy locations, we adapt the existing methods for mapping of uncertain geodata. For missing geographical location, we propose a tentative visualization of relative positions between physical

27 Dream Cartography Concepts

dream elements, but also of the relative position between dream characters and physical elements (e.g. near the door, in a corner).

A special case of settings in dreams is condensation of two or more places. When features of two or more different, familiar objects are combined to create a new object in dream, we speak of creative condensation (Blechner, 2013, p. 62). The condensation of places, or, for that matter, of persons in dream is a known phenomenon, described already by Freud (1942). In the nine- teenth century Galton is known to have merged two photographs together to form a composite image. This process, achieved with computers, is nowadays called “morphing” and we choose this technique to represent the condensation of places in dream. Morphing is mostly used in film industry, but also in criminal police for faces of suspects. Takeuchi and Perlin illustrate the application of morphing to city design (2012). There are also other techniques such as puzzle- like collage (Goferman, Tal, & Zelnik-Manor, 2010) of maps or the simple depiction of puzzle tiles on the unchanged world map, which are currently under consideration for condensed or mixed places. Although morphing is seen nowadays mostly as a dynamic process of smoothly transforming one image into another one, we emphasize here the smooth transition made between map parts as different from a map collage. The output of morphing in this case is a map with harmoniously combined elements, as it would represent a continuous region and which is congruous with the transitions experienced in dreams. In contrast, in a puzzle-like collage the individuality of the compositing elements of the collection is not only clearly visible but also, sometimes, even purposely highlighted.

For the non-spatial dream content, we propose alternative visualization models such as diagram-like representations of time flow and character occurrence and graph- or network- like representation of the social interactions, the latest inspired for example from Moretti’s plot analysis (2013). For the coding of dream elements’ values (e.g. types of social interactions with the properties of the appearing characters, the valence/emotions, the existing of goal directed behaviors, the attitude of the dreamer) we combine established visual variables: color hue, color value, shape, size, orientation, texture (Bertin, 1983), color saturation (MacEachren, 1994) and transparency (Wilkinson, 1999); use topology, as a link to geography, and specifically created symbols in maps and diagrams or networks.

Part of the visualization concept is the user participation, therefore the display of a first rough visualization draft on the screen as soon as possible, followed by an interactive editing phase: enter of additional information or adjusting of the visual elements. Participants may thus try alternative offered visualizations to get one that “feels” right for them, and so participate to the map composition for obtaining a more detailed and meaningful dream map. A good reason for this participation is to avoid beforehand interpretation of the dream in the dream map. Finally, brings a richer understanding of the dream itself because of the possibility to choose elements from a superordinate category (Montangero, 2009) for depicting dream places, actions or objects. In this way Dream Cartography lets participants visually reformulate their dreams. We

28 Paper I

illustrate hereafter as proof-of-concept some of our proposed visualization elements for dream space, respective for social encounters in dream. 6 Visualization Examples For the examples below we refer to openly available dream reports, in particular to the dream series published in German by a philosophical psychologist and university professor in book form (von Uslar, 2003). Von Uslar wrote down and documented more than 6000 of his own dreams between 1949 and 2001. Two examples are given, one to illustrate the possibilities of a dream setting and one to show social encounters in a dream. 6.1 Dream Space Here a fragment of a dream report, written by Von Uslar in 1999 in Zurich (translation from German by C.M. Iosifescu Enescu):

I lived in a university town, which bore combined features of Göttingen and Zurich. From Göt- tingen it had the old town, but from Zurich the buildings of ETH and the University, which lie above the city. In addition, both parties had long been separated by the Iron Curtain between East and West. I had already lived for decades down in the old town, which resembled Göttin- gen, and now I was again, for some reason, heading up to the ETH.

Here, the dreamer shapes the setting from the beginning of the dream re-port and this turns out to be also the most significant element of the studied dream scene. With respect to the dream elements and their characteristics stated in Table 3, we understand the dream location to be an area of walking distance, part of a middle-sized city, formed of components from of two real cities, therefore a condensation of places occurred (see section 5.3).

29 Dream Cartography Concepts

Figure 1 Morphing example of cities of Göttingen and Zurich, which appear as a condensed place in a dream; Original images: Bing Maps, Bird’s Eye Both original cities are familiar to the dreamer. Figure 1 and Figure 3 propose visualization elements for the dream setting: maps created by morphing / combination of elements from the two cities in the aerial view, respective at the city plan level. For their creation, we made use of the textual information from the dream report and the world knowledge, respective the actual location of the old city in Göttingen and of the universities buildings in Zurich.

Adding available background information about the dreamer and his written awake thoughts connected to the dream, we learn that the person studied in Göttingen (almost 50 years previous to the dream) and in the last 30 years was working (teaching) at the University of Zurich. To-

30 Paper I

gether with the mention of time in the dream report, “for decades”, this signalizes another prominent element of this dream: the time flow. Figure 2 shows the living locations of the dreamer in his real life, as they succeeded by time.

Figure 2 Dreamer's living locations in time; in red the “Iron Curtain”, which separated East from the West Germany before 1990; Data: Google Maps & scribblemaps.com The dreamer spent his youth years in a town from East-Germany (Schwerin), which was separated from the West-Germany by the “Iron Curtain”, mentioned also in the dream report. Thereby the city of Schwerin is also, as the dreamer himself writes down, part of the “background” setting. We visualize the third town in the city plan (Figure 3a), by using a red transparent mask over the topographically higher part (see also Figure 3b) of the new city (Zurich part) in red, constant with the meaning of the red color in Figure 2. Also in red we draw the delimitation line between the parts of the two cities, with a spatial interruption and a path where the dreamer could have walked through in his dream.

31 Dream Cartography Concepts

Figure 3 City plan resulted from the combination of the old city Göttingen (upper left) with Zurich universities (bottom right) with components delimitated from each-other with a red boundary line (a) and with elevation information (b); Original data sources: city plans: OpenStreetMap contributors6; elevation: Nasa SRTM7 Figure 3b uses a shaded relief for conveying the height information, which is addressed in the dream. The shaded relief was generated from the digital elevation model obtained by combining the elevation models of the two cities.

The next example (Figure 4) shows another feasible visualization for a dream setting, which uses both an absolute position and relative positions. This is a fragment of a dream report, written by Von Uslar in 2001 in Zurich (translation from German by C.M. Iosifescu Enescu):

6 http://www.openstreetmap.org, accessed on 30.10.2014 7 http://www2.jpl.nasa.gov/srtm/, accessed on 11.11.2014

32 Paper I

I lived in a house that was right on the edge of a rocky slope, which went hundreds of meters into the depth just in front of the house. Below, directly at the foot of the cliff, stood the Bellevue Palace in Potsdam from which one could see directly on to the rock wall. […]

Figure 4 3D visualization for a dream referring both absolute and relative positions, including keyword objects; Original data sources: map: OpenStreetMap contributors; cliff: parts from different Magritte paintings and an own generated elevation model; stone house: hookedonhouses.net8; 3D model of the Sanssouci Palace: MaZach, 3dwarehouse.sketchup.com9; Perspec- tive visualization in Terrain Bender Along to the dream, Von Uslar writes his thoughts at wakeup and there we learn that he mis- takenly calls in his dream the palace in Potsdam as Belvedere instead of Sanssouci, because of its “beautiful view” (which is the meaning of Belvedere in Italian) to the stone wall. Moreover,

8 http://hookedonhouses.net/2011/04/10/for-sale-a-small-stone-house-in-new-hope-pa/, accessed on 15.04.2015 9 https://3dwarehouse.sketchup.com/model.html?id=100ed196ab746f081290dc43fc303fc, accessed on 15.04.2015

33 Dream Cartography Concepts

he references a Magritte painting, which represents a stone house on a floating cliff (which we found to be “The Castle of the Pyrenees” by René Magritte). These expressive visual dream elements ask for a visual dream depiction. On that account, we grab keywords from the dreamer’s text, such as: stone house, cliff, Magritte painting, Palace Sanssouci, and we proceed to an image search over the Internet. There are nowadays many publicly available picture li- braries (e.g. Flickr10, social media), which can be automatically searched for a keyword and deliver usable pictures on any subject. An algorithm for graphical salience, for example, could make a first selection of some pictures, which are to be further presented to the user or randomly chosen from and put together in a collage. The collage, which finally represents the dream scene, must, of course, respect the given relative positions (“on the edge of”, “below”, “at the foot of”) of the keyword elements.

The beauty of this dream is that it not only contains strong keywords and relative positions, but also an absolute position, a geographical place (Potsdam, Germany). The combination of the previously made collage with a perspective view of the map, including artistic deformation (performed for this example with the Terrain Bender11) empowers the dream scene with a location factor and a further visual element, in this case resulting in a new, fascinating landscape composition. 6.2 Social Encounters Based on another dream fragment from Von Uslar (translation from German by C.M. Iosifescu Enescu) we illustrate one of our endeavors toward the visualization of social interactions in a dream:

[…] There I come together and talk with two boys, one of which I find very pleasant. […] I ask him how can I reach “Grossen Moor” street and if a Mr. von Hartwig lives there. […] I’m asking about other acquaintances of my father as well, because I don’t know where I could spend the night. [...] He sees it as a matter of course that I will spend the night at his place [...] and before that he would have to ask his foster father.

10 https://www.flickr.com/, accessed on 15.04.2015 11 http://www.terraincartography.com/terrainbender/, accessed on 15.04.2015

34 Paper I

The dreamer encounters two persons in this dream scene and with one of them he develops a conversation. A continuous, thicker line with arrowheads on both sides symbolizes this relationship (Figure 5). Other persons are just mentioned in the conversation and they appear connected with interrupted line to the present / other mentioned persons.

Figure 5 Example for the depiction of the social interaction network; “I” stands for the dreamer’s person Color values of the profiles symbolize emotions or attitudes of the dreamer towards the other persons: we chose green for sympathy and yellow for the father as relative. No recognizable attitude we coded as grey. The color saturation signifies in this example the participation of the persons to the dream scene. The network nodes that contain the characters may vary on the visual variable shape. Circles were chosen here (Figure 5) for the overall friendliness of the dream scene and other shapes (e.g. convex or concave) could stand for other emotional load.

The topological arrangement of the network is also an information carrier, relative edge length signifying the perceived relation closeness between dream actors, with or without actual interaction (edge materialization). The dreamer stands in the center of the network and the appearance of the persons in the dream scene is coded basically clock-wise, with nodes pulled out of this order when the drawing of connection-edges requires it.

We sketched human profiles (Figure 6 shows the masculine profiles) for different ages, considering the change in physiognomy with the age and we use these for the representation of persons in the social encounters network. If the age of a person plays a role in the dream scene, these profiles may be used accordingly (e.g. for “boys” in the Figure 5).

35 Dream Cartography Concepts

Figure 6 Proposed aging masculine profiles to be used for characters in the social interactions network of a dream Several such series along with feminine profiles and indefinite gender profiles complete our collection and present a choice for the visualization of the actors in the social interactions network in dreams. The alternative is to use real pictures of the dreamed persons and cover these with semi-transparent color masks, but this depends on the willingness of the participants to upload pictures and requires pictures with a neutral face expression, for not intervening with the overall dream emotion (e.g. a picture with sad face expression into a happy dream scene).

We mention that other diagram and network styles are under consideration for the visualization of the social interactions network of a dream, and separate visualization propositions will be made for character occurrence in dream series. 7 Conclusion and Future Work Dream Cartography is part of the humanized geography movement, along with the Literary Geography and other mapping systems, where the human being is explicitly taken into account and where experiences, emotions, and individual preferences or interests make a difference in the resulting map. In Dream Cartography we propose a visual “reformulation” of dreams based on dream reports, additional thoughts about the dream and common sources. In this paper we address in particular the data acquisition issue, stressing that there are ways to improve the quality of the dream reports, for example, by following instructions that guide the process of dream logging. Moreover, valuable information may be gathered through additional questionnaires.

Regarding the visualization elements, a key concept is the interaction of the dreamer or of the psychologist with the cartographic system. Therefore, the final step of data modelling, as well as the different map elements, may be refined by the users themselves, with the system proposing different visualization methods and map compositions and the participant choosing in an interactive mode the best-suited map result. This takes place online, on a specially prepared Web platform. Special effects such as themes (e.g., a treasure map or child’s sketch) applied to location maps and to the other visualization elements can increase the visual appeal of the dream map.

The vast complexity of the dream worlds suggests the need for a variety of visualization methods and thus innovative requirements for Dream Cartography, which we are looking forward to addressing in our future work.

36 Paper I

8 References Baylor, George W., and Corrado Cavallero. 2001. “Memory Sources Associated with REM and NREM Dream Reports throughout the Night: A New Look at the Data.” Sleep: Journal of Sleep Research & Sleep Medicine. Bertin, Jacques. 1983. Semiology of Graphics. University of Wisconsin Press. Blechner, Mark J. 2013. The Dream Frontier. Routledge. Cipolli, Carlo, and Diego Poli. 1992. “Story Structure in Verbal Reports of Mental Sleep Ex- perience after Awakening in REM Sleep.” Sleep: Journal of Sleep Research & Sleep Medicine. Cresswell, Tim. 2013. Place: A Short Introduction. John Wiley & Sons. Dennerlein, Katrin. 2009. Narratologie des Raumes. Walter de Gruyter. Desseilles, Martin, Thien Thanh Dang-Vu, Virginie Sterpenich, and Sophie Schwartz. 2011. “Cognitive and Emotional Processes during Dreaming: A Neuroimaging View.” Con- sciousness and Cognition, From Dreams to Psychosis: A European Science Foundation Exploratory Workshop, 20 (4): 998–1008. Domhoff, G. William. 1996. Finding Meaning in Dreams a Quantitative Approach. Emotions, Personality, and Psychotherapy. New York, NY: Plenum Press. ———. 1999. “New Directions in the Study of Dream Content Using the Hall and Van de Castle Coding System.” Dreaming 9 (2-3): 115. Domhoff, G. William, and Adam Schneider. 2013. “DreamBank.” http://www.dreambank.net/. Egenhofer, Max, and David Mark. 1995. “Naive Geography.” Spatial Information Theory A Theoretical Basis for GIS, 1–15. Forshage, Mattias. 2007. Drömgeografi, naturgeografi. Styx. Foulkes, David. 1999. Children’s Dreaming and the Development of Consciousness. Harvard University Press. Frank, Andrew U., David Mark, and Martin Raubal. 2013. “Researching Cognitive and Lin- guistic Aspects of Geographic Space: Las Navas Then and Now.” In Cognitive and Lin- guistic Aspects of Geographic Space, edited by Martin Raubal, David M. Mark, and Andrew U. Frank, 1–22. Lecture Notes in Geoinformation and Cartography. Springer Berlin Heidelberg. Freud, Sigmund. 1900. Die Traumdeutung. Goferman, Stas, Ayellet Tal, and Lihi Zelnik-Manor. 2010. “Puzzle-like Collage.” In Computer Graphics Forum, 29:459–68. Wiley Online Library. http://onlineli- brary.wiley.com/doi/10.1111/j.1467-8659.2009.01615.x/full. Hall, Calvin Springer, and Robert L. Van de Castle. 1966. The Content Analysis of Dreams. New York: Appleton-Century-Crofts. Hau, Stephan. 2006. “Traumforschung.” In Psychoanalyse Und Psychiatrie, edited by Heinz Böker, 359–69. Heidelberg: Springer Medizin Verlag. Hurni, Lorenz. 2008. “Multimedia Atlas Information Systems.” In Encyclopedia of GIS, 759– 63. Berlin, Heidelberg: Springer.

37 Dream Cartography Concepts

Iosifescu Enescu, Cristina M., Ionut Iosifescu Enescu, and Lorenz Hurni. 2013. “Customizable Map Content and 3D Visualization in the Online Platform GeoVITe.” In Proceedings of the 26th International Cartographic Conference. Dresden, Germany. Iosifescu Enescu, Cristina M., Ionut Iosifescu Enescu, Helen Jenny, and Lorenz Hurni. 2011. “GeoVITe - a Service-Driven Solution for an on-Demand, User-Friendly Web Access to Geodata.” In Proceedings of the 25th International Cartographic Conference. Paris, France. Iosifescu Enescu, Ionut. 2011. “Cartographic Web Services.” Dissertation for the degree of Doctor of Sciences, ETH Zurich. Kerr, N. H. 1993. “Mental Imagery, Dreams, and Perception.” In Dreaming as Cognition, edited by C. Cavallero and D. Foulkes. England: Harvester Wheatsheaf. Kriz, Karel, William Cartwright, and Lorenz Hurni. 2010. Mapping Different Geographies. Lecture Notes in Geoinformation and Cartography. Berlin: Springer. MacEachren, Alan M. 1992. “Visualizing Uncertain Information.” Cartographic Perspectives 13: 10–19. ———. 1994. “Some Truth with Maps: A Primer on Design and Symbolization.” Association of American Geographers, Washington. MacEachren, Alan M., Anthony Robinson, Susan Hopper, Steven Gardner, Robert Murray, Mark Gahegan, and Elisabeth Hetzler. 2005. “Visualizing Geospatial Information Un- certainty: What We Know and What We Need to Know.” Cartography and Geographic Information Science 32 (3): 139–60. Montangero, Jacques. 1999. Rêve et cognition. Sprimont: Mardaga. ———. 2009. “Using Dreams in Cognitive Behavioral Psychotherapy: Theory, Method, and Examples.” Dreaming 19 (4): 239–54. doi:10.1037/a0017613. ———. 2012a. “Dreams Are Narrative Simulations of Autobiographical Episodes, Not Stories or Scripts: A Review.” Dreaming 22 (3): 157–72. ———. 2012b. “Cognitive Approach to Dreaming.” In Encyclopedia of Sleep and Dreams, edited by Deirdre Barrett and Patrick McNamara Ph.D. Santa Barbara, California: ABC- CLIO. Moretti, Franco. 2013. Distant Reading. Verso Books. Nir, Yuval, and Giulio Tononi. 2010. “Dreaming and the Brain: From Phenomenology to Neu- rophysiology.” Trends in Cognitive Sciences 14 (2): 88–100. Peterson, Michael P. 2008. International Perspectives on Maps and the Internet. Lecture Notes in Geoinformation and Cartography. Berlin Heidelberg New York: Springer. Piatti, Barbara, Hans Rudolf Bär, Anne-Kathrin Reuschel, Lorenz Hurni, and William Cart- wright. 2009. “Mapping Literature: Towards a Geography of Fiction.” In Cartography and Art, edited by William Cartwright, Georg Gartner, and A. Lehn, 177–92. Lecture Notes in Geoinformation and Cartography. Berlin Heidelberg: Springer.

38 Paper I

Reuschel, Anne-Kathrin. 2012. “A Literary Atlas of Europe (Poster) : Ein Literarischer Atlas Europas.” http://www.literaturatlas.eu/files/2012/03/Literary_Atlas_of_Eu- rope_POSTER_gut.pdf. Reuschel, Anne-Kathrin, and Lorenz Hurni. 2011. “Mapping Literature: Visualisation of Spa- tial Uncertainty in Fiction.” The Cartographic Journal 48 (4): 293–308. Reuschel, Anne-Kathrin, Barbara Piatti, and Lorenz Hurni. 2013. “Modelling Uncertain Geo- data for the Literary Atlas of Europe.” In Understanding Different Geographies, edited by Karel Kriz, William Cartwright, and Michaela Kinberger, 135–57. Lecture Notes in Geoinformation and Cartography. Berlin Heidelberg: Springer. Strauch, Inge, and Barbara Meier. 1996. In Search of Dreams: Results of Experimental Dream Research. SUNY Press. Takeuchi, Yuichiro, and Ken Perlin. 2012. “ClayVision: The (Elastic) Image of the City.” In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2411–20. CHI ’12. New York, NY, USA: ACM. Tenbrink, Thora, and Werner Kuhn. 2011. “A Model of Spatial Reference Frames in Lan- guage.” Spatial Information Theory, 371–90. “The Landscape of Our Dreams - Foundations of Dream Geography.” 2011. http://www.surre- alistgruppen.org/the_landscape_dreams.pdf. Tuan, Yi-Fu. 1977. Space and Place: The Perspective of Experience. U of Minnesota Press. Von Uslar, Detlev. 2003. Tagebuch des Unbewussten : Abenteuer im Reich der Träume. Würz- burg: Königshausen & Neumann. Wilkinson, Leland. 1999. The Grammar of Graphics. Statistics and Computing. Springer-Ver- lag, Inc., New York.

39 Dream Cartography Concepts

Paper II. Fictional volunteered geographic information in Dream Cartography This is the Author’s Original Manuscript (AOM) of an article published by Taylor & Francis in International Journal of Cartography in March 2017

Cristina M. Iosifescu Enescu, Lorenz Hurni

ETH Zurich, Institute of Cartography and Geoinformation, Zurich, Switzerland

Abstract: In the field of fictional and literary geography, the spatial information is conveyed through the power of words by one person: the author of the novel or, in the case of Dream Cartography, by the dreamer. Every story takes place somewhere. However, the acquisition of geographic information differs conceptually from the geography of the real, physical world. This represents a novel opportunity for engaging in the production of volunteered geoinformation. With the right tools available, the collaborative spatialization of dreams gives birth to a new category of Volunteered Geographical Information (VGI), namely a fictional VGI. We argue that not only the dreamer, but also a third party may contribute to fictional geography by making a visual interpretation of the original content published by the dreamer. The social aspect of volunteering is discussed in both VGI and F-VGI. Additional possible sources of information, which may be used for retrieving geoinformation from dreams, are also mentioned. Our judgements are supported by data from an experimental study on dream locations. The participants rated their awareness of the dream setting, stated which elements are helping them to recognize places in their dreams and estimated the frequency of their dreams happening in personally relevant places.

Keywords: Fictional VGI (F-VGI), Dream Cartography, Fictional Geography 1 Introduction Referencing to geographic information is deeply embedded in everyday tasks, applied, for example, when people talk about their leisure activities such as a trip to the countryside, when

40 Paper II

searching for a restaurant to have dinner, or when giving directions. Drawing a schema to ex- plain to the spouse where the new grocery shop is located is also a way of creating and sharing geographic information. However, this is not usually perceived as such. The same happens when people recount a dream: the dream takes place somewhere; and sometimes it would be nice to be able to better describe this place, to remember it or to “visit” it another time. Dream Cartography proposes to do just that, using specially developed tools.

The goal of Dream Cartography is to create dream maps, with the focus on the locations where the dreams take place and proposing the visualization of other dream elements as well (C. M. Iosifescu Enescu, Montangero, & Hurni, 2015). Although usually not a central element in a dream, the dream setting is important. It is an anchor, which helps people to build a story out of the dream, to make sense. If the dream setting is remembered, people tend to begin a dream report by giving information about it first (e.g., on dreams reports saved by Domhoff & Schnei- der, 2013).

We speak about fictional places, such as in literature or dreams. However, one could argue that the dream settings are not fictional, or at least not consciously intended to be fictional. Yet one must differentiate here between fiction and fantasy (which is a subset of fiction). Moreover, we take into consideration the Reality Assumption: “everything that is (really) true is also fiction- ally the case, unless excluded by the work” (Friend, 2017, p. 29). This renders dream reports to be fictional works, as they relate about a dream world, where people also “have arms and legs” (Friend, 2017, p. 32), if not otherwise mentioned. Returning to places, by implication, a setting in a fictional work can be called fictional.

The aim of the current work is to show how a new form of volunteered geographic information (VGI) is about to arise, namely a fictional-VGI (F-VGI), where people compose maps of fictional worlds, such as exist in dreams or in novels. The particularities of the F-VGI compared to the VGI are addressed, focusing on aspects such as volunteers or sources of data. Our judgements are supported by the results from an experimental study on dream settings. 2 State of the art Goodchild (2007, p. 211) designated Volunteered Geographic Information (VGI) as the harnessing of Web tools “to create, assemble, and disseminate geographic information provided voluntarily by individuals”. Few would assess fictional work as a carrier of geoinformation. However, every story needs a place, where the events “take place”, and therefore “it’s impossible to even think literature without any spatial context” (Piatti et al., 2009, p. 178). This space and its composition is the subject of literary geography, one of the fictional geographies. A fictional text such as a novel or a dream report is indeed a source of geoinformation. While the literary geography deals with settings in literature, there are some important differences when trying to transfer their findings to Dream Cartography; the main ones are the length of the original text (the dream report) and the fading memory of the dream (C. M. Iosifescu Enescu et al., 2015).

41 Dream Cartography Concepts

Although VGI is fundamentally based on the assumption that contributions come to a conver- gence and, thus, over time, the differences between multiple versions of reality smooth over (Elwood, Goodchild, & Sui, 2012), this is not necessarily the case in F-VGI. Multiple versions, multiple visual interpretations of the same fictional text are allowed.

Due to the length of a dream report compared to the length of a literary text, but also due to the particularities of the dream space, it is almost impossible to apply automatic methods for GIR to dreams. Although different methods have been developed for automatic retrieval of a place name or description from an unstructured text (Jones & Purves, 2008; Jones, Purves, Clough, & Joho, 2008; Schockaert, De Cock, Cornelis, & Kerre, 2008), these are using additional information in the studied text, which are rarely available in a dream report. Moreover, GIR could hardly account for sudden transitions from a place to another, which are very far away in real world; in dreams, the distance between these may be very different from the real one. As a dream-blogger writes about his dream of going by car from the USA to Japan: “Please, don’t ask me how we made it through deep blue sea by car. I don’t know neither.” (Tougaw, 2009, p. 253). Rather, and because we are talking about volunteers, the geoinformation retrieval must still be manually performed by the dreamer herself or by a third party. As Haklay, Singleton and Parker (2008, p. 2026) put it: “Crowdsourcing is how large groups of users can perform functions which are difficult to automate or expensive to implement”.

A literary text is a complete work; on the contrary, a dream report is “only” a description of the dream. Although the dream reports are a good measure for dreams (Nir & Tononi, 2010) and are almost exclusively the means used by the dream research community, they are far from being complete. Previous work on Dream Cartography (C. M. Iosifescu Enescu et al., 2015) shows how to best get the written report of a dream and its possible sources listed by the dreamer herself.

Another issue in fictional geography is the scale of space, in which the story is happening, and which resemble to the psychological spaces. Montello (1993) classifies the psychological spaces into four classes: figural, vista, environmental and geographical. Figural space is smaller than the human body and is the space of pictures, maps, objects, or distant landmarks. Vista spaces can be visually apprehended from a single point of view, such as a single room, a town square, the horizon, or even the surface of the earth viewed from above. Environmental space is larger than the human body and surrounds it, its apprehension requires “the integration of information over significant periods of time” (Montello, 1993, p. 315) and is a constructive process. Examples of environmental places are buildings, neighborhoods and cities. Geograph- ical space (states, countries) is so large that it cannot be apprehended directly, but learned through symbolic representation, such as maps or models.

42 Paper II

2.1 Volunteers Volunteers in GI “are highly likely to be living in an advanced economy and to be a member of the middle class, thus to have the education, technical skills, access to resources and infrastructure that facilitates participation in these activities.” (Haklay, 2013, p. 112). Making a parallel to our study on dream locations (see section 3) we notice how good the gathered convenience sample resemble in the age and education structure to the volunteers presumed to be the most active in the VGI.

Many people show a strong interest in their dreams. In spite of an assumed difficulty in collecting data for the Dream Cartography project (especially because of privacy issues), many sources of dream data were found. Openly available dream reports are found in published books and dream diaries (e.g., von Uslar, 2003; Adaman-Tremblay, 2015), on dream forums (klartraum.de, dreamviews.com, dreamscloud.com) or on scientific online dream databases (dreambank.net). This happens although dreams are very personal experiences and they reveal much of the current concerns and activities of the dreamer, as the continuity hypothesis of dreaming and its experimental basis sustain (Strauch & Meier, 1996; Domhoff, 1996; Erlacher & Schredl, 2004).

As Goodchild (2007) does for VGI, we must also wonder what drives the people to be accurate. And the same answers are valid for dream data: self-promotion, audience, but what may add to these is the wish to have an ‘interpretation’ of their dream. This is highly plausible, given the fact that there are platforms just for dream interpretation on the Internet (dreamscloud.com, or even the Facebook page of the International Associations of Dreams). There are also many people, more or less ‘expert’, willing to give their interpretation to a dream. Interesting discussions develop, to which the dreamer is fully participating, by giving new information, confirm- ing or not the interpretations. For example, the motto of dreamscloud is “We make dreams social” (dreamscloud.com, 2015). In this regard, revealing dreams can be compared to the use of social platforms such as Facebook, where the desire of self-revealing and self-disclosure reaches its apogee (Hollenbaugh & Ferris, 2014).

Just as in VGI, where local activities and changes in local infrastructure or names are best known and fast documented by local people (Goodchild, 2007), the F-VGI is best documented by the dreamers themselves and in a short time after the dream. The dreamers themselves, psychologists, artists: everyone can participate, be an F-VGI volunteer, and create a dream map. The volunteers extract the geographical information from the dream / dream report and, using available tools (e.g. on a Dream Cartography platform), they assemble, visualize and disseminate it. Third parties contributing to a dream visualization may be commissioned by the dreamer or might be acting from their own initiative on open dream data. In the latter case, a direct interaction with the dreamer is not given. Therefore, other sources could help to complete the lack of data.

43 Dream Cartography Concepts

2.2 Data and information sources In the case of F-VGI, the possibility to go in the field and measure physical proportions is not provided, since the original source is a description of the setting in a dream report (or, in the case of literary geography, in a fictional text). Luckily, it handles about the same, written, text. To the information extracted from the original text, one may add other sources. This situation has some similarities with traditional VGI, where it is possible to vectorize sources of secondary information, but also major differences.

Fictional settings are not measurable because they vary in their relationship to the real geography. They can be extremely realistic and have a real-world counterpart, but can also be a cross- fading of more real spaces or even partially or totally invented (C. M. Iosifescu Enescu et al., 2015). Reuschel and Hurni (2011, p. 294) suggest that “the spatial dimension in fiction [..] has to be completed and developed through the imagination of the reader”. Generic world knowledge on history or geography helps to further develop the “imagination”, but an important source of information is the author’s biography, especially in the case of dreams.

A key variable in investigating the memory sources of dreams is the temporal delay to which an element from the waking life is incorporated into a dream. The most elements in dreams are from the day (already Freud (1942) introduced the concept of “day-residue”) or the week (recency effect) prior to the dream (Grenier et al., 2005). It is therefore important that, besides a dream report, the dreamer writes down also what she has been doing the day before, and lists possible sources of her dream. Here an example, on how the day-residue can affect the dream content:

Back in the eighties, I was allowed to watch together with my parents the American prime time television soap opera “Dallas”. An episode was broadcasted each Sun- day evening and I, as a child, used to go to bed just after the show. The action was catchy and it happened to me quite a few times that, when the film started next Sunday, I was surprised to see where it begins: I used to dream my personal continuation of the series. In my dreams, I was either one or the other of the main characters and experienced everything from that perspective. My continuation had, of course, little in common with the real following episode and I’ve never been in Texas in my real life. [personal experience of the first author]

Identifying the dreamer’s biography as an additional information source, we are interested in the weight of the temporal extent of biography elements occurring in dreams. Grenier and colleagues (2005) found in dreams references to autobiographical memories from the whole life span. The participants to their experimental study (30 older women, >60 years, from 217 dreams) identified temporal references in the own dreams and, separately, produced a sample of autobiographical memories. The authors found a linear decrease in temporal references identified in dreams and autobiographical memories with increased remoteness for the last 30 years

44 Paper II

(Grenier et al., 2005). For older participants, they could also show the occurrence of a reminiscence bump (here the age from 10 to 19, known from the research on autobiographical memory as important for identity formation, see also Conway & Haque, 1999). Figure 1 illustrates these results, with the mention that, for a clearer visualization, the authors excluded from the graphic the last category of “more than 60”, which reflects the recency effect and contains 65% of the memory and 85% of the dream data.

Figure 1 Distribution of memories and dreams in terms of a person’s age at the time of the original experience; after Grenier et al (2005, p. 286).

The results of Grenier and colleagues (2005) show that features of past memories (e.g., old residence sites, personally relevant geolocations) may be incorporated for a longer period of time into dreams. This is also what showed our experimental study (C. M. Iosifescu Enescu, 2016) on dream locations (see section 3). 2.3 From VGI to F-VGI Maps and information available on open platforms, which originate form VGI, are used in many domains. Even for Dream Cartography, geographical information may be downloaded from one or the other of these platforms and combined or morphed in such a way (Figure 2) that it best fits the space description in a dream (for the resulting dream visualization, see (C. M. Iosifescu Enescu et al., 2015).

A good example of F-VGI are maps of the territories from certain books or TV-series, which enjoy a lot of attention from the community. One of the best (Figure 3) is made by fans for the book series “A song of ice and fire” or TV series “Game of Thrones”, and presents a very detailed, interactive map with character plots and trajectories structured by book chapters or TV-series episodes; it is licensed under the Creative Common license (theMountainGoat & Tear, 2012). The mapmakers combined elements from different drawings and from different parts of the Earth in satellite view from Google Maps. Moreover, we learn that a Web forum, “The Cartographers’ Guild”, is ready to take any commissions for maps for fictional realms, as

45 Dream Cartography Concepts

used both in novels and computer, tabletop or role-playing games and we may admire various creations of these authors on their site (http://www.cartographersguild.com).

Figure 2 The schema, on how a morphed city was composed from two city plans, of Zurich (top) and Göttingen (bottom); Original source data: OpenStreetMaps contributors Remembering the discussion about world knowledge as data source, as we may notice in Figure 3, the map of Westeros resembles the geography of Ireland, with the North to South (see “the Fingers” compared to region Cork – Tralee in Ireland). Moreover, the author G. RR Martin seems to have followed the relief and historic sites of Ireland (this time not turned to 90 degrees, as the map itself) in building his cities, castles and story course on the map. Let us consider for example Dublin, which in Gaelic means “black pool”. Dublin influenced the “Blackwater Bay” from King’s Landing, the capital of Westeros, which is, similarly, situated in the Middle East of the country (shakeyspears, 2014).

46 Paper II

Figure 3 The map of the fictional territory Westeros (left) compared with the map of Ireland (right); Sources: http://quartermaester.info/ and Google Maps There are many resources available for inspiration when creating fictional maps. These and other personal experiences may become part of a dream, as we have discussed. In the following, we show how people perceive their dream settings based on a self-conducted experimental study on dream location. 3 A study on dream locations We studied the dream location and different dream characteristics using a mainly self-developed questionnaire. The questionnaire was available for six months (July 2015 – January 2016) and was completed by 255 people (172 females, 83 men), the most with a high education and aged between 20 and 45. The questionnaire was online on an external server (soscisurvey.de) and was offered in four languages: English, German, French and Romanian. Active publicity was made for the questionnaire through personal E-Mails, posts on dream-research communities Web pages and Facebook posts. It resulted a convenience sample (for details on methods for selecting research participants, see (Gravetter & Forzano, 2003). The questionnaire asked people, among others, to estimate the frequency of their remembered dreams in the last couple of months and, on this base, to answer some questions about dream location and sleep and dream patterns. Part of a larger project (C. M. Iosifescu Enescu, 2016), the target group of this study consisted of people with international migration background. The questionnaire ad-

47 Dream Cartography Concepts

dressed also the question of dream location related to real locations. The participants were instructed to think about their dreams, recall where the dreams were happening and answer if they recognized the dream setting. Due to its length, complexity and specific questions related to international migration, many people did not finish the questionnaire. However, for general questions about dream location (see Figure 4Fehler! Verweisquelle konnte nicht gefunden werden.), we could analyze 309 answers.

Regarding the interesting point of how people recognize a place in their dreams (Figure 4), only 16% recognize the place due to landmarks, 31% due to environmental elements (a river, a forest, building styles, etc.), 64% just have the feeling to be there and 13% do not recognize a place at all. Moreover, the results show a correlation between the reported visual memory and the remembering of the dream setting: People, who try to remember details of the landscape when they are in a new place, are more aware of the location in their dreams.

In the case of the international migration and how this influences the dream settings, the Dream Location Questionnaire focused on the smallest scale: at the country level. People were asked, in how many percent of their dreams appeared specific countries, which were of interest to them. All of the interviewed people stated they do dream about countries, which were current, past or future residences, holiday countries or heritage culture. However, when asked about the percent, they gave values, which rarely added up to 100%, being both more or less than that. We account for this fact with the explanation that this is specific to dreams: People may dream in one dream about both their childhood country and their current residence country (leading to mixed or condensed places) or, on the contrary, their dream location may have less with the real world in common, but consisting of imaginary places.

Figure 4 Dream Location Questionnaire, page with general questions on dream location. Analyzing the results from a cross-sectional perspective, we found that the age at the time of the last migration significantly influences the amount of dreams about the current residence

48 Paper II

country. People who moved before they turned 19 years old to the current country of residence dreamed more about this country than people who were older when they moved to their current residence country. The interpretation of this result (C. M. Iosifescu Enescu, 2016, p. 42), is that the current residence country (although different from the one they were born in) might represent for these people a second “childhood country”. This is because enough memories point to this country, also from a time where the identity formation takes place. This period is shown to occupy a relatively big amount of memories also in the autobiographical memory, being called the reminiscence bump (Rubin, Wetzler, & Nebes, 1986), as previously discussed.

Table 1 Examples of dream locations given by people in the dream location survey

Places with a real-world coun- Mixed places or unclear Fantasy places terpart Childhood and current home A semi-familiar garden; an In a cafeteria, which doesn’t University, school unknown balcony exist Workplace, office In or around water A class room in an apartment In a hotel room I’ve lived in A tree-house in a city center of a city un- At people's houses - people Random houses known to me important to me My high-school, only darker, School (but not a school I The summerhouse of my par- more sinister, bigger have ever been to in real life) ents Woods; meadow Underground tunnels filled At a theatre I’ve been to As a pedestrian on a road with water in an unknown lo- Holidays resort I’ve been to In a train cation on the world map. Outside, at a nice place Pubs similar to those I know, A kind of technological vor- where I’ve been to streets similar to those I tex Places to which I just was in know but not really any exact Beach / holidays resort, excursion place. More of a similar im- where I've never been Childhood neighborhood pression of the place. A strange small airport Town center of the town An overlap between two dis- A spaceship where I now live parate city suburbs; Random In the Wild West (after I mashed-up locations of a city played a computer game I know I dream about the city about it) I live in, but it looks different The topic / setting of a book I in my dreams, a fantasy read or of a film I’ve seen town, which is much more Fictional city densely built and is darker. My dream world, from which Europe, shrunk into a land I I write fiction could walk about and visit different cities.

49 Dream Cartography Concepts

Table 1 shows a few examples of places people named as their dream settings in the last couple of months, categorized by their relationship to a real place and ordered by the scale of the space. People mentioned mostly well-known spaces such as at work or at home. These were experienced usually in large scales which, translated to the psychological spaces of Montello (1993) as introduced in chapter 2, correspond mostly to vista or environmental scale, rather than to geographical or figural scale. 4 Conclusions and outlook The new concept of fictional VGI was described considering important aspects such as the sources of data and who are the volunteers. We mentioned also some differences between F- VGI for Literary Geography and F-VGI for Dream Cartography. The social aspect was touched, where the volunteers are the dreamers themselves, engaged in a quest of self-discovery and, at the same time, of self-revealing. However, a third party can perform as well a dream visualization based only on a dream report and using additional sources of information. Dream sources are shown to be highly connected to the dreamer’s experiences from the previous day (day- residue) or to the dreamer’s biography. Therefore, personally relevant geolocations are a good starting point for visualizing dreams, where the dream setting is too vague. Dream locations are experienced mostly from a psychological scale of space called vista, and this aspect was illustrated with examples from the Dream Location Questionnaire. Our study shows that people, which moved to another country, rated their dreams to be happening more often in that residence country, where they lived at the age of identity formation, at the beginning of adulthood. Compared to a dream report, a literary text is usually longer and therefore presents a bigger probability to contain enough data for geoinformation extraction. VGI and F-VGI resemble in the aspects of having a main data source to which additional sources of information may be added; however, these are conceptually different: measurable versus non-measurable, just as the spaces themselves can be: linear versus nonlinear.

In conclusion, volunteering is a good way to approach the acquisition of geoinformation from an unstructured text. F-VGI is well suited for Dream Cartography, especially when the volunteers are the dreamers themselves. With the right tools available, individuals will be able to represent their dream worlds or allow others to do it for them and vote for the quality and accuracy of interpretation or for the artistic depiction of their dreamscape. Regarding the tools for Dream Cartography, more research is needed to reveal the form these should take and how exactly the scale and the context of space can be taken into account, in order to map the dream settings in an expressive manner. 5 References Adaman-Tremblay, L. (2015). My Night Life. Lidia Adaman-Tremblay, Lulu.com. Retrieved from http://www.lulu.com/shop/lidia-adaman-tremblay/my-night-life/paperback/product-21982102.html

50 Paper II

Conway, M. A., & Haque, S. (1999). Overshadowing the Reminiscence Bump: Memories of a Struggle for Independence. Journal of Adult Development, 6(1), 35–44. Domhoff, G. W. (1996). Finding meaning in dreams a quantitative approach. New York, NY: Plenum Press. Domhoff, G. W., & Schneider, A. (2013). DreamBank. Retrieved January 25, 2016, from http://www.dreambank.net/ dreamscloud.com. (2015). DreamsCloud | We Make Dreams Social. Retrieved October 21, 2015, from https://www.dreamscloud.com/en/about Elwood, S., Goodchild, M. F., & Sui, D. Z. (2012). Researching Volunteered Geographic In- formation: Spatial Data, Geographic Research, and New Social Practice. Annals of the Association of American Geographers, 102(3), 571–590. https://doi.org/10.1080/00045608.2011.595657 Erlacher, D., & Schredl, M. (2004). Dreams reflecting waking sport activities: A comparison of sport and psychology students. International Journal of Sport Psychology, 35(4), 301–308. Freud, S. (1942). [1900] Die Traumdeutung. (A. Freud, Ed.) (Vol. Gesammelte Werke II/III). London: Imago. Friend, S. (2017). The Real Foundation of Fictional Worlds. Australasian Journal of Philoso- phy, 95(1), 29–42. https://doi.org/10.1080/00048402.2016.1149736 Goodchild, M. F. (2007). Citizens as sensors: the world of volunteered geography. GeoJournal, 69(4), 211–221. https://doi.org/10.1007/s10708-007-9111-y Gravetter, F. J., & Forzano, L.-A. B. (2003). Research methods for the behavioral sciences. Toronto, Canada: Thomson. Grenier, J., Cappeliez, P., St-Onge, M., Vachon, J., Vinette, S., Roussy, F., … De Koninck, J. (2005). Temporal references in dreams and autobiographical memory. Memory & Cog- nition, 33(2), 280–288. Haklay, M. (2013). Citizen Science and Volunteered Geographic Information: Overview and Typology of Participation. In D. Sui, S. Elwood, & M. F. Goodchild (Eds.), Crowdsourcing geographic knowledge (pp. 105–122). Netherlands: Springer Science & Business Media. Retrieved from http://link.springer.com/chapter/10.1007/978-94-007- 4587-2_7 Haklay, M., Singleton, A., & Parker, C. (2008). Web mapping 2.0: The neogeography of the GeoWeb. Geography Compass, 2(6), 2011–2039. Hollenbaugh, E. E., & Ferris, A. L. (2014). Facebook self-disclosure: Examining the role of traits, social cohesion, and motives. Computers in Human Behavior, 30, 50–58. Iosifescu Enescu, C. M. (2016). Impact of Migration on the Dream Setting (Master’s thesis for the degree of Master of Science in Psychology UZH). University of Zurich, Zurich, Switzerland. Iosifescu Enescu, C. M., Montangero, J., & Hurni, L. (2015). Toward Dream Cartography: Mapping Dream Space and Content. Cartographica, (50.4), 224–237.

51 Dream Cartography Concepts

Jones, C. B., & Purves, R. S. (2008). Geographical information retrieval. International Journal of Geographical Information Science, 22(3), 219–228. Jones, C. B., Purves, R. S., Clough, P. D., & Joho, H. (2008). Modelling vague places with knowledge from the Web. International Journal of Geographical Information Science, 22(10), 1045–1065. Montello, D. R. (1993). Scale and multiple psychologies of space. In European Conference on Spatial Information Theory (pp. 312–321). Springer. Retrieved from http://link.springer.com/chapter/10.1007/3-540-57207-4_21 Nir, Y., & Tononi, G. (2010). Dreaming and the brain: from phenomenology to neurophysiol- ogy. Trends in Cognitive Sciences, 14(2), 88–100. Piatti, B., Bär, H. R., Reuschel, A.-K., Hurni, L., & Cartwright, W. (2009). Mapping literature: towards a geography of fiction. In W. Cartwright, G. Gartner, & A. Lehn (Eds.), Car- tography and Art (pp. 177–192). Berlin Heidelberg: Springer. Reuschel, A.-K., & Hurni, L. (2011). Mapping Literature: Visualisation of Spatial Uncertainty in Fiction. The Cartographic Journal, 48(4), 293–308. Rubin, D. C., Wetzler, S. E., & Nebes, R. D. (1986). Autobiographical memory across the lifespan. In D. C. Rubin (Ed.), Autobiographical memory (pp. 202–221). Cambridge: Cambridge University Press. Schockaert, S., De Cock, M., Cornelis, C., & Kerre, E. E. (2008). Fuzzy region connection calculus: Representing vague topological information. International Journal of Approx- imate Reasoning, 48(1), 314–331. shakeyspears. (2014). (No Spoilers) Westeros is Ireland: A detailed analysis of how Ireland inspired the creation of Westeros : asoiaf. Retrieved September 25, 2016, from https://www.reddit.com/r/asoiaf/comments/1zglmr/no_spoilers_westeros_is_ire- land_a_detailed/ Strauch, I., & Meier, B. (1996). In search of dreams: Results of experimental dream research. New York, NY: SUNY Press. theMountainGoat, & Tear. (2012, February). A Song of Ice and Fire Speculative World Map. Retrieved from http://quartermaester.info/ Tougaw, J. (2009). Dream Bloggers Invent the University. Computers and Composition, 26(4), 251–268. https://doi.org/10.1016/j.compcom.2009.08.002 von Uslar, D. (2003). Tagebuch des Unbewussten : Abenteuer im Reich der Träume. Würzburg: Königshausen & Neumann.

52 Paper II

53 Dream Cartography Concepts

Paper III. Place Cookies and Setting Spiders in Dream Cartography This is the Author’s Original Manuscript (AOM) of an article accepted for publication in the journal Transactions in GIS 2019, special edition on Modelling and Analyzing Platial Repre- sentations

Cristina M. Iosifescu Enescu*, Hans-Rudolf Bär*, Matthias Beilstein**, Lorenz Hurni*

*ETH Zurich, Switzerland, Institute of Cartography and Geoinformation

**Beilstein Cartographic Services, Switzerland

Abstract: This paper contributes to understanding the difference between objective space and subjective place. New data models and visual methods, which make possible the comparison between dream settings, are necessary to an exploratory analysis of dreams. The subjective perception of settings is decomposed by studying dream reports, by applying a survey and considering related scientific literature. This leads to the construction of two data models, which are applied in dream cartography. The place cookie model features the dreamer’s familiarity with the setting, being visualized in the form of concentric circles. The setting spider model is based on 26 variables, extensively characterizing the setting. These are grouped into eight factors, and visualized in a compact radar chart with eight “legs”. As a superordinate system of the setting spider, the event spider is developed, describing the whole dream scene. The proposed models and visualization methods can be transferred for real life events (settings).

Keywords: Dream Cartography, Place Cookie, Setting Spider, Place vs. Space, Event Spider, Place Metadata Profile 1 Introduction In dreams, it is not always easy to identify the place of the action. There are different elements from the surroundings, which may possibly play a role hereof: the natural environment, such as a river, a forest edge or a hill; or human-made structural elements such as buildings, specific

54 Paper III

landmarks, roads or architectural style. When dreaming about indoor locations, is there the fur- niture, the wall decorations, the arrangement and dimensions of the rooms, which disclose the location?

The challenge of dream cartography is to visually represent dream locations, where dreams take place and, most importantly, the characteristics of these places. Dream cartography is an interdisciplinary research area between cartography and psychology, which has the goal to map the dream worlds of volunteering dreamers for an exploratory analysis of dreams. An exploratory analysis implies discovering and highlighting (unexpected) patterns, if there are any. In order to discover patterns in dream locations, there is a need for an informed overview of a dream series (many dreams of a single person); or of spatially or thematically related dreams of different persons. This can be achieved only by being able to compare dream locations.

Yet not all dream locations have a geographical reference, which might be a problem for creating maps. A solution is to focus rather on the qualities of a place, than on its exact position in the world. In dream research, the term “setting” is used for referring to dream locations, which comprises both the location and the ambience, being similar to a theatrical scenery. A starting point is to observe how people describe their dream settings. When dream settings are known, corresponding to real locations, their description resemble closely to that of a real place, as expected from a visitor of this location in awaken state. This is obviously true because people describe dream locations using their experience in describing real-life places. In this context, it is critical how people perceive a place, in the absence of external media means of registering.

Moreover, it is acknowledged that “the simple location of the events alone is not sufficient to grasp the meaning associated with place” (Caquard & Cartwright, 2014, p. 102). Consequently, the question is: What makes places distinctive and how can these be meaningfully compared to each other? The approach presented in this work consists of decomposing a setting into its subjective characteristics, into its qualities for an individual. This serves the purpose of making places (or dream settings) comparable with each other. The resulting models have to be understood as the best effort, and do not restrict the existence of other possible models.

In the following, previous research inspiring this work is summarized, and then the currently employed methods are presented. The paradigm of dream cartography is used to develop models and visual representations of a setting based on only one of its characteristics: the “place cookie”; or on a multitude: the “setting spider”. Finally, the models and their visualizations are applied to dream setting analysis, for a comparison between different dream settings. 2 State of the Art The subjectivity in the perception of space became a debate in the Geographic Information Science (GIS) community lately, where more and more people collect data and information that transform space into place. Place is loaded with personal meaning, place is the space directly experienced by the people (Tuan, 1977; Tilley, 1994). Platial data refers to place-based data,

55 Dream Cartography Concepts

rather that space-based (Westerholt, Mocnik, & Zipf, 2018). Dream settings are par excellence platial, as exemplified in Iosifescu Enescu & Hurni (2017).

Dennerlein (2009) analyzes in detail how locations are described in literary works. The author identifies how space is referenced: as toponyms (e.g. Berlin, Atlantis), with proper names (e.g. Dracula’s Castle, Boeing 747), by class designation (e.g. dining room, inner city), deictically (here, there) and with other concrete words (e.g. east, west, inside, outside, foreign, afar, dark- ness). Similarly, place is referenced deictically, contextually (e.g. on the right of, on the left of, here, there) or absolute. Furthermore, the absolute place can be: intrinsic (e.g. in the front of the house), topologically (e.g. in, on, next to), geographic (e.g. at the equator, south of Frankfurt, on Rosenthaler street) and metric (e.g. hundred meters further).

On the other hand, in dreams, literary science and storytelling, the concept of “setting” is used when referring to place. Steele (1981, p. 11) defines the setting as “a person’s immediate surroundings, including both physical and social elements”. So it comprises, besides the location, also its atmosphere. It is this combination of elements that makes a place exciting, worth of being the stage of a narrative. 2.1 Elements of a Setting Individuals have their own “axes-system”, the human space is centered on the person and therefore places are qualitatively different for different people (Bollnow, 1963). Based on these qualities, a personal classification of places comes into play, which has no counterpart in the abstract, mathematical space.

Dream research that looks into settings usually only reports on whether this place is familiar or not, inside or outside, geographical or not (Domhoff, 1996; Strauch & Meier, 1996). Previous research in dream cartography (Iosifescu Enescu, Montangero, & Hurni, 2015, p. 229) proposes a questionnaire for a dream scene (event in a dream). This comprises some questions, which are most relevant for characterizing the dream setting, e.g., the clarity of representation (“clearly distinct, vague or fuzzy”), the landscape and prototypical genre (“built-up area, country-side, hills, mountains, seaside, exotic, fantasy or science fiction”) or the senses used in the dream (“vision, hearing, touch and smell”). In literary cartography, Reuschel and Hurni (2011, p. 296) address two crucial elements describing the setting: the relation between the literary space and the real space (“imported, transformed, invented, imagined, synthesized or shifted places”) and the function and meaning of literary space (“simple scenery, thematic scenery, protagonistical physical, protagonistical psychological, mythical connotation and symbolical connotation”).

Foucault (1967/2008) mentions other characteristics of space that are of interest, such as: private versus public space or cultural versus useful space. Lefebvre (1974/1991) argues that the (social) space is a social product, and that space embodies social relationships. Therefore, important elements characterizing a setting are the culture, or the spoken language, but also a possible contradiction between the place and the circumstances. In a globalized world, not only the culture can make a difference, but also the wealth, the income: e.g., high income classes of

56 Paper III

different cultures may possibly have more related standards and views than high- and low- income classes of the same culture (Rosling Rönnlund, 2016). Moreover, Montello (1993) identified different scales of experiencing space: when space is smaller than the human body, such as distant landmarks or maps; or when the space is larger than the human body and surrounds it.

Bernstein and Roberts (1995) developed a questionnaire for investigating dream content. Some of its questions are related to time (the dreams generally taking place in the present, the past or the future) and setting: remaining constant or changing in a dream, or being “restrictive and confining” or “spacious and vast”. However, whether time is part of the setting or not, is not clear. A dreamer would determine the time period of a dream not necessarily during the dream, but rather after the awakening (J. Montangero, personal communication, December 12th, 2018), based on other elements such as place familiarity, culture or congruence / expectancy.

The Golledge’s anchor points theory studies the formation of cognitive maps and proposes anchor points as basic elements of spatial knowledge representation. These give a structure to the cognitive space, organizing it hierarchically and regionally. Anchor points may be any physical feature of the environment, which is salient for an individual, corresponds to a routine place, is frequently interacted with, or has significance in a person’s life. This corresponds to a “multi- dimensional measure of familiarity” and therefore places like home or work place are anchor- points “in every person’s mental map” (Couclelis, Golledge, Gale, & Tobler, 1987, p. 105).

Indeed, the familiarity of a place to an individual is a key variable in the perception of space. This is illustrated also in narratology, where the concept of familiar versus unfamiliar settings is well studied, being dealt with in fundamental works such as Joseph Campbell’s comparative mythology or in Yuri Lotman’s “Structure of the Artistic Text”. Therefore, the archetypal hero (Campbell, 1949/2008) must cross a physical boundary between a known, familiar space (e.g. home) to an unknown, mysterious space (e.g. forest). Lotman (1971/1977) uses this even for classifying literary works: with plot, where the crossing of such a boundary happens, and plotless (not boundary-crossing).

Regarding dreams in general, and not specifically the setting, there exists comprehensive analyzing and coding systems for dreams, which are used in dream psychology. The Hall/van de Castle system (Hall & Van de Castle, 1966; Domhoff, 2003) includes the following elements: characters, social interactions, activities, success and failure, misfortune and good fortune, emotions, settings, objects and descriptive elements. However, the category emotions is too detailed. Russell (1980) introduces a model for reducing affective states to two dimensions: arousal (numbing to intense) and valence (negative to positive); this model is successfully applied, e.g., for mapping space-related emotions out of user-generated photo metadata (Hauthal & Burghardt, 2013).

57 Dream Cartography Concepts

In the following two subsections, the sources of inspiration for the models and visualization developed in the current work are described. The social network (section 2.2) influenced the creation of the place cookie model, due to the similar variable used: the familiarity, and to the simple visualization. The political profile (section 2.3) influenced the creation of the setting spider model, due to the meaningful grouping of a multitude of variables and their compact representation. 2.2 The Social Network The two models of the social network shortly described here use different terminology, but both similarly classify the social relationships of a person. Research in sociology and anthropology (Dunbar, 1993, 1998, 2010; Hill & Dunbar, 2003) determines the size of the social network in humans to be in average 150 persons (the Dunbar’s number), to which people maintain personal contact. Four typical hierarchically inclusive levels of personal relationships are modeled depending on the frequency of contact (Figure 1A), therefore on the intensity of the relation individuals maintain with their peers. The smallest group is the support clique (contact once a week), followed and included by the sympathy group (contact once a month), followed and included by the close network (to whom it requires a conscious effort to keep in contact), which is finally followed and included by the cognitive group (or personal network). Dunbar (2010) reveals the expected size of these groups, in number of persons: roughly 5-15-50-150 (therefore increasing in size by a multiple of three). The classification can continue to the supernetwork (500) or the language community (1500). In his book “How many friends does one person need?” Dunbar refers to these grouping levels as circles of acquaintanceship (Dunbar, 2010, p. 33).

Another model of the social relationships is called “circle of friends” or “circle of support” (Falvey, Forest, Pearpoint, & Rosenberg, 1997), and evolved from a different perspective. The goal of this model is to create awareness and actively enhance the number of people in an inner circle, e.g., in case of illness (Pearpoint, 1991). This approach features as well four concentric circles (Figure 1B). Innermost is the circle of intimacy and comprises the people to whom an individual has intimate relationships: the partner and the close family. This is followed and included by the circle of friendship (regular contact), followed and included by the circle of participation (people seen occasionally, with shared interests), which is finally followed and included by the circle of exchange (paid relationships). An important difference between the two approaches is that, whereas by Dunbar the number of people contained in a circle is already given in average, by Falvey et al. the number of people contained in the two middle circles may be actively increased in order to achieve more support, become integrated into community and influence personal wellbeing.

58 Paper III

Figure 1 Visually summarized uses of concentric circles for classifying the social network of a person. (A) Hierarchically inclusive levels of acquaintanceship by Dunbar (2010); (B) Circles of support by Falvey et al. (1997) Concentric circles are a compact, self-explanatory visualization for the variation of one factor: in this case, the familiarity. However, it can be that more factors influence the perception of an element, be it setting, social relation or political belief. A valuable method for the classification and visualization of the influence of multiple factors on one element, the political profile of an individual, is presented in the following. 2.3 Political Profile or the Smart Spider Hermann and Leuthold (2003, 2001) model the political views based on factor analysis, for the “Atlas of Political Landscape in Switzerland”. The variables describing the political views are grouped into factors. These are wisely paired and set on opposite sides of an axis, e.g. horizontal axis (economic dimension) with its left (social welfare) versus its right (fiscal conservatism) factors and vertical axis (cultural dimension) with its liberal (internationalism) versus conserva- tive (anti-immigration) views (Figure 2). Due to their project “sotomo”, the authors are well known in Switzerland for the creation of the “smart spider”, a radar chart for visualizing political profiles of public persons or of parties. With the sotomo project, a voting advice application was built, which offeres Swiss citizens the possibility to determine their own political profile by answering a set of questions, in order to facilitate their voting choice. This application is also adapted for the European Union (Trechsel, 2011); however, the Swiss version features 8 factors on 4 axes, whereas the other has only 7 factors, but no axes: the circle-chart is divided as a pie with a uneven number of slices. The advantage of having axes (dimensions) on the radar-chart, and not only simple factors, is straightforward: an axis can host rather opposite characteristics and more load of one side of the axis could imply less load on the other. The load in this case

59 Dream Cartography Concepts

(from inner to outer circle) represents the summarized agreement to a factor. This is a good match between content and visualization technique, providing instantly a clear overview of the situation.

Figure 2 Smart-spider diagram for Swiss elections 2009 (Hermann, 2010) Two of the data models developed in this paper, the setting spider and the event spider, are similar to the smart spider. 3 Methods The goal is to develop a data model, which can be used for setting comparison. In a first step, an initial list of questions related to space, place and atmosphere in dreams was created based on consulted scientific literature on dream psychology, cartography, geography and narratology (see section 2.1).

In a second step, random dream reports were analyzed on their platial dimension, and answers to these questions were given or estimated by the authors. These dream reports were selected from the well-known online dream reports repository dreambank.net, both in English and Ger- man, with the criterion of having between 150 and 300 words. The selection was done by using the random sampling provided online by the repository. All studied dream reports were written down by the dreamers themselves in their own dream diary for personal use and were only afterwards donated or published for scientific purposes. The temporal lag between dreaming and writing the dream down is usually in these cases a couple of hours (in the morning after the dream).

The purpose was to verify the validity of the questions and, if needed, to update the list with new questions, to best reflect the observed specificity of a setting. Some questions, which turned

60 Paper III

out to be irrelevant for most of the studied dream reports were removed. Similar questions were consolidated. After this step, from the initial list 17 questions remained and they were considered to sufficiently describe the dream setting.

One of these questions seemed to be especially important, because it could be answered in most of the studied dream reports: “Are you familiar with the setting?”. This was indeed expected, based on the studied literature. The dreamer’s familiarity with a place is therefore analyzed in detail: using the example from the social network (section 2.2), the place cookie model (or the personal circles of places) is created (see section 4.1).

In a third step, the suitability of the 17 questions to characterize the setting was investigated by conducting a paper-and-pencil survey. The purpose was to verify if the list was complete, gather potential suggestions for new items, and discover possible ways of grouping the questions. The participants were 30 cartography and GIS specialists from the own academic research group (25 to 60 years old, 7 women), having a good understanding of space. They answered the questions regarding their own (dreaming) experiences.

The questions were named (e.g., congruence) and numbered, and will be addressed in the following as setting components. The following 17 components were evaluated in the survey: familiarity, routine, time, relation to geospace, role of space, accessibility, spiritual/practical, congruence (expectancy), attention, emotional impact, population density, space volume, light, temperature, complexity, clarity of representation and scale (perspective). An additional line, with number 18, was left empty for the participants to fill up, if they felt that the setting was not sufficiently described by the 17 questions. For each component, single choice questions / statements were prepared, followed by three to five options (compare to Table 3, for the final list of components), e.g., for congruence:

To me, this setting… was in line with the happenings was neither in line nor in contradiction with the happenings was in contradiction with the happenings

The answer options are modeled and ordered from usual to unusual. The terms “usual” and “unusual” in this case are used as they are understood in the world of consciousness and rational logic as a baseline. However, not all components fit soundly in this scheme. An extreme case is accessibility: being a private space or a public space. Whereas this could be solved by asking, in addition, which is the most usual form of space, public or private, this does not meet the purpose, because being private versus being public is a stand-alone characteristic of a place. Where it is not clear, which answer is more usual, that answer is taken, which is most probable. For this it is considered the perception of a setting in waking state, because dreams often reflect waking concerns and therefore the real life of the dreamers, as the continuity hypothesis of dreams states (Domhoff, 1996; Strauch & Meier, 1996; Schredl, 2003; Erlacher & Schredl, 2004).

61 Dream Cartography Concepts

Table 1 Overview of the survey Task I: Task II: Dream Autobiographical Memory Tasks Please think about a dream Please recall a vivid biograph- you remember vividly. This ical memory, preferably dream had one or several set- something from your child- tings. Put an A to each of the hood. Put an X to each of the statements, to which you statements, to which you agree. agree. No. of components 17+1 17+1 No. of returned ques- 26 26 tionnaires No. of filled question- 22 22 naires Maximal no. of answers 374 (22*17) 374 (22*17) No. of missing answers 4 (1%) 28 (7%)

There were two tasks on this survey, therefore the participants responded twice to the 17(+1) questions. The first task required the participants to recall one of their dreams and the second task to recall a vivid biographical memory (see Table 1). Participants’ were given about 10 minutes to fill out the questionnaire and were instructed to answer the questions intuitively, without long deliberation. The participants were not asked to write down their dreams or memories. The tasks were conceived to make a person concentrate on specific aspects of a dream. In order to prevent people to consciously or unconsciously add, modify, or interpret dreams to fit expectations, and because the single-choice questions are straight-forward, the time for answering the questionnaire was kept short. The limited time functioned also as an incentive, to convince more people to participate.

Form 26 returned questionnaires, 4 participants did not respond to the first task, stating they do not remember any dream, but responded to the second task. 4 (other) participants did not respond to the second task (but responded to the first task), due to time constraints (it took them longer to remember a dream, as the discussion afterward cleared). The returned questionnaires were anonymous. Verbal feedback from the participants confirmed that the questions were clear and understandable, and were perceived as relevant for both tasks, with two major exceptions: the question related to time for both tasks, and the question related to geospace for the second task.

62 Paper III

Regarding the time of the dream or of the biographical memory, the answer options were:

in the current time in the own past in the near future in historic times in the far future no defined time

For the dream task, many participants felt that it was difficult to situate a dream into a time period. For the autobiographical memory task, participants did not see the reason to be asked about time: a biographical (childhood) memory takes always place “in the own past”. This question is less suited for describing the setting.

However, time plays an important role in a GIS for places, as for example an assessment of a place can occur in the current time (time point being registered), in recollection (a known past time usually determined by an event), or in imagination (in the absence, in the unknown past, current or future time). Time is not necessarily related to the setting, but is a separate, important attribute of the event or dream scene. Therefore, time was removed from the proposed setting components and was promoted to an upper hierarchical level: the elements characterizing the event, so that the event (or dream scene) spider model was additionally created (see section 4.3).

The question regarding the relation of the autobiographical memory’s setting with the geospace (second task, first question) was ignored by most participants. A discussion after the survey revealed that for them it was obvious that, for a memory, the only correct answer is “geographical”, and thus this question is not suitable for a memory. However, memories can be influenced by media (Kligler-Vilenchik, Tsfati, & Meyers, 2014), and therefore also a memory’s setting can be transformed and become mixed or even imagined.

The main goal of the survey was to see how well suited are the questions, and this was assessed, beside the discussion with the participants afterwards, by comparing the maximal possible number of answers with the number of not answered questions (see Table 1). For the dream task, there are only a few missing answers (1%), so that the questions seem to be well-suited to describe the dream setting.

However, for the autobiographical, there are 7% missing answers, not counting the 4 unfilled questionnaires. Some of the missing answers (9) can be explained by the time constraint (the last questions of the second task were not answered). Moreover, answers were systematically missing for the component “relation to geospace”, which was previously discussed. Other missing answers for task two can be explained by a failure in the exact wording of the answer options (e.g. for the role of space: “the setting is replaceable without any consequences to the dream”, overlooking to replace the word “dream” with “happenings”). Even if missing answers can be

63 Dream Cartography Concepts

explained, the questions seem to be more suited for describing a dream setting than for describing a memory’s setting. However, a confounding variable can be the expectancy of the participants, because the survey was announced as being about dream settings, and only when ex- plaining the tasks the autobiographical memory was mentioned.

Another use of the survey was to see how the components relate to each other, in order to categorize them, if possible. This is how the decision was taken to group familiarity together with routine. There were significant (p <0.01) correlations of 42% between the answers of a place being familiar and being a routine, respectively of 58% between the answers of a place being unfamiliar and not a routine, for both, combined, tasks. However, there were also other correlations observed in the data, which cannot be explained by the overlapping of the variables. For example, for the participants, a routine place in a dream setting was usually a restrictive place (Pearson correlation of 61%, p<0.01) whereas a no-routine place was spacious (59%).

Some of the returned questionnaires featured a new question or setting component at number 18. The participants made following suggestions: whether the setting is natural or artificial, landscape configuration (altitude differences) and noise. Other participants described shortly their dream afterwards, and other missing elements crystalized, such as weather conditions, connectivity to other places, or whether the physical laws are different (e.g. flying in dreams, or being weightlessness on an orbital space station).

Finally, finding the last setting components and grouping them into factors, implied another iteration over the existing literature, and more discussions held with different cartography (from the own research group) and psychology specialists. A total of 26 setting components resulted. 4 Data Models In the following, two models of the setting and a model of the event (or dream scene) are presented. Initially developed as a proof-of-concept, the place cookie turned out to be a stable, independent model. This is illustrated below. Subsequently, the more complex model, the setting spider, is explained. At the end of this section, the event spider model is introduced, which arose from the desire of an overall assessment of a dream scene (or of an event). 4.1 The Place Cookie Model Analogous to the concentric circles of the social network, the knowledge of space is modeled in circles of intimacy. This model is called the “place cookie”, nicknamed after its associated visualization (Figure 3), but also after its characteristic of saving personal relevant information (compared to Web cookies).

In the social relationships, the intimacy decreases and group size increases towards the outer circle. Applying this reasoning to places, the very familiar places are situated in the inner circle and the less familiar places in the outer circles; outside of the place cookie remain the unknown places. Very familiar places are for example the current home, or childhood places – places to which the individual has close, intense relationships. Less familiar, but still well-known are,

64 Paper III

e.g., the work place, the way to work or to school, friends’ homes or other places visited often. In the third circle, of known places, there are, e.g., holiday destinations or other places, which have been seen only once, or did not occupy the memory, the attention for a long time. Places individuals only learned about, but have never been to, are situated in the outer circle. To pin- point a place onto the place cookie, an ordinal rating scale is proposed. The individuals answer one question (Table 2), rating their familiarity or intimacy to a place on a scale from 1 (very familiar) to 6 (totally unfamiliar). Each number on this scale has also a qualitative description (e.g., intimate, such as home), to make the rating easier and make sure the answers of different individuals can be compared in the end.

Table 2 Ordinal-scale for determining the familiarity to a location How do you rate your relationship to this place? (from 1: very close – to 6: no relationship) 1 Intimate (e.g. home) 2 Well known (e.g. the work place or school class, the way to work or to school) 3 Known (places you often go to, e.g. a nearby park, the fitness studio) 4 Seen (e.g. a one-time holiday destination) 5 Learned about (read in a book, seen in pictures, maps, etc.) 6 Unknown – outside the concentric circles

On the place cookie, it is possible to visualize the familiarity of a single place, but also the platial difference between two places. Yet the rated familiarity is an ordinal measure and the intervals between the circles are not necessarily equal. Therefore comparisons (e.g., a place is more familiar than another) are allowed; differences, however (e.g., the difference in familiarity is 2 points), must be handled with caution. Figure 3 shows places which can be familiar to an individual (from 1 to 6) and, as a bonus, their emotional valence to this person: positive (with +) or negative (with -), but also neutral (with =).

65 Dream Cartography Concepts

Figure 3 Example of a filled place cookie The place cookie is the precursor of the complex model of the setting, the setting spider (or the setting metadata profile). 4.2 The Setting Spider Model This model is based on 26 setting components. Some of them correlate to each other, as previously discussed (familiarity and routine) and yet others belong together, such as temperature and light, making out the environmental conditions. Therefore, the components are grouped into factors, similar to the model of the voting advice application in Switzerland (see section 2.3). The 26 components of a setting are listed on the outside of the circle in Figure 4, and detailed in Table 3, together with their grouping into factors, respectively their division into human or physical characteristics of setting.

Some of the components contribute with a greater weight to the value of their factor, because they are perceived as more important (in dreams): familiarity to the factor individual exposure, scale to the factor morphology and physical laws to the factor environment. The components’ weight to the factor composition and an overview of the answer options are indicated in Table 3. The exact single-choice statements for each component and their values are listed in Appen- dix. Appendix B.

66 Paper III

Table 3 Setting factors and their components

Factors Components Scale Weight (usual-unusual) (%) Relation to Relation to geospace geographical / mixed / imagined 100 geospace

few people / very crowded or de- Population density 20 serted Wealth same / richer / poorer 20 Society Language same / known / unknown 20 Culture same / known / unknown 20 Spiritual/Practical practical, useful / spiritual, magical 20 Familiarity familiar / unfamiliar 60 Individual

Human characteristics Human Routine routine / no routine 20 exposure Prototypical place prototypical / not prototypical 20 Attention received attention / active moment 33 Clarity of representa- Attention blurred / very clear or indistinct 33 tion Congruence in line -/ in contradiction with action 33 Role of Role of space replaceable / physical influence 100 space

Scale normal -/ higher -/ lower perspective 33 Space volume not sure / restrictive / spacious 17 Morphology Landscape/Curvature no -/ high altitude differences 17 Complexity low complexity / very complex 17 Neatness not perceived / very neat or messy 17 Accessibility private / public / not sure 33 General Ex- Connectivity well - / not connected 33 posure Natural/Artificial both / artificial / natural 33 Physical characteristics Physical Physical laws respected / different 40 Weather not perceived / extreme 15 Environ- Temperature not perceived / extreme 15 ment Light not perceived / extreme 15 Sound conversation / very noisy or silent 15

The setting components were grouped into eight factors, which are visualized on a radar chart with four axes, or eight radii – “legs” – therefore the name of spider diagram. The most important two components regarding the specificity of the setting in dreams formed one factor each. These are situated on the vertical axis: the relation of a place to a physical, geographical

67 Dream Cartography Concepts

location (top); respectively the influence of the place on the happenings, stating if the scenery becomes a protagonist or not (bottom, see Figure 4).

Figure 4 The setting spider or place metadata profile: defining factors for a subjective perception of a setting and their components This vertical axis divides the spider diagram into its right half, which hosts characteristics of the settings related to human geography; and its left half, which features physical characteristics of settings. In the right half, the individual exposure to a setting is formed of three components: the individual’s familiarity with the setting; whether the setting is routine or not and whether the setting is prototypical or not. The factor society contains components describing the setting as it is affected by other people: population density, wealth, language, culture, being a practical or a spiritual place. The attention, a factor depending on the person and the momentarily circumstances, comprises the component with the same name (attention), but also the clarity of representation and the congruence.

68 Paper III

In the left half of the setting spider, the general exposure is formed of the accessibility (public or private place), the connectivity of a place to others and whether the setting is rather artificial or natural. The morphology, the more strictly physical characteristics of a setting, comprises the scale, to which the setting is perceived, the volume, the landscape or curvature, the complexity and the neatness of the setting. The environment, which depends on the momentarily circumstances, draws on the physical laws, weather, temperature, light and sound conditions of the setting.

Moreover, the factors are organized by pairs, which are placed on opposite sides of the chart on the same axis. The pairs have opposing character (human vs. physical), but also content similarities. On the horizontal axis are the individual versus the general exposure factors, and on the two secondary axes the factors society versus morphology, respectively the factors attention versus environment (see Figure 4).

The usual-unusual scale means that, if a dream setting nearly corresponds to usual conditions, its profile forms a polygon with the smallest area on the radar chart (close to the center). The default value of all questions is set to “usual”. On the contrary, if a dream setting is unusual, extreme or bizarre, its profile forms a polygon with a larger area on the chart. Therefore, the bigger the area of the formed polygon is on the setting spider, the more unusual is the dream setting.

Moreover, a polygon pointing to the left of the circle indicates more unusual elements in the physical characteristics of the setting (e.g. scale / perspective, altitude differences, accessibility, weather). In contrast, a polygon oriented more to the right represents a bigger load of unusual elements in the human characteristics of settings (e.g. familiarity, culture, wealth, attention, expectancy). 4.3 From the Setting to the Entire Event: the Event Spider Model Location is paramount in cartography. However, for dream research, there are also other elements of dreams, which are at least as important as settings. The previous radar-chart approach with eight opposing factors is also applied for dreams in general (see Figure 5 and Table 4), using the categories proposed in previous dream research (see section 2.1). One factor of the event spider model is the setting itself (highlighted in Table 4). It can be considered that the setting spider model is a detailed description of this one factor, and shall be build (only) if the setting is important in a dream, in an event.

69 Dream Cartography Concepts

Table 4 Event (or dream scene) factors Factors / Compo- Scale nents (not important-important) Time no - / current - / known - (past) / unknown time (future) active, Actors not important / important / very important struggle, Goal not important / important / very important living Emotions: Valence not relevant / positive / negative Setting not important / important / very important passive, Objects not important / important / very important effortless, Fortune not important / important / very important inanimate Emotions: Arousal weak / intense

This means that, when negative emotions are present in a dream, the profile polygon has a bigger area on the event / dream spider, than with positive emotions.

Figure 5 The dream or event spider, metadata profile for a dream / an event The factors of the event spider model are not composed of many components, they can be answered as one question per factor. The setting is positioned on the event spider as a factor paired

70 Paper III

on the same axis with the time. The other pairs are: actors versus objects, goal versus fortune, and emotions valence versus emotions arousal. Analogous to the setting spider, the factors were assigned to the two halves of the circle; left: the passive, the effortless, the inanimate versus right: the active, the struggle, the living (Figure 5). 5 Application in Dream Cartography The two models of the setting developed previously are applied for visualizing the dream setting in dream cartography. First, it is illustrated how painting a ring of the concentric circles of a place cookie can transform it in a univariate point symbol on a map. Second, the setting spider is applied on a personal dreamland map of a test person. The setting spiders become as well point symbols on the map, in this case multivariate point symbols. Other possible uses of the two models are also mentioned. 5.1 Place Cookies In the description of a dream, the setting may be referenced by the place name, e.g. “Helve- tiaplatz”, by its significance to the dreamer, e.g. “a place where I pass by with the bus every working-day and never stop”, or both at the same time. When this happens, there are two elements describing the setting: the proper name, which can be geocoded, and the familiarity of the place, which can be visualized on the place cookie. The advantage is that the place cookie (in this case familiarity 4) can be placed as a point symbol (Figure 6 left) on the city map. If the dreamer mentions as well her office and its location (e.g. ETH Zurich main building) in the same or in another dream, the place cookie with the office’s familiarity value (e.g. 2, Figure 6 right) can be on the same map, at its geocoded location. The place cookie is used as a univariate point symbol. This results in a map (Figure 6), where both the spatial distance between the two places and the platial difference of these places for the dreamer can be represented, in terms of familiarity, using place cookies. In this example, the well-known OpenStreetMap representation was especially chosen as base map, stressing that this approach may be easily used by anyone, with any map, and also for real places (not only for dream locations).

71 Dream Cartography Concepts

Figure 6 Map using the place cookie as point symbol for spatial locations, Background data: OpenStreetMaps contributors Another application of place cookies in dream cartography is in visualizing dream series. For example, an aggregation of all dream settings on only one place cookie is possible, illustrating the occurrence of places with different familiarity in a person's dreams. The evolution of a place regarding its familiarity in a dream series could also be displayed on a place cookie. Moreover, using place cookies, specific places in dreams of different individuals can be compared (by their familiarity): noticing, for example, that the nightmares of the person A take place more often in unknown than in familiar locations, whereas those of the person B take place usually in very familiar locations. 5.2 Setting Spiders Similar to the place cookie, it is essential to test the setting spiders. While for the place cookie this was straightforward (only one component), for setting spiders it is not that simple, because of the many components of the setting / questions, which need to be answered. For this, a test person J. answered the 26 questions for a couple of his dreams (dream reports recorded in his personal dream diary), in order to determine the values of the setting components and calculate the factors’ values. The factors were mapped on the setting spiders and the idea was to put these charts on a map.

J. dreams very vividly and about one third of the dreams that he remembers take place at geographical or mixed geographical locations. He is an expert in cartography and, finding out about the dream cartography project, he offered his help and produced a beautiful and artistic map of his dreamland using about 40 of his dreams, written in a dream journal. The setting spiders are placed on this map (Figure 7).

72 Paper III

J. made artistic choices for the map creation. He started with the mountains, because, for a Swiss person, the relief is one of the most important elements on the map. He physically modeled small mountains out of play dough and photographed them. This is how he obtained the shaded relief, which looks very convincing on his map. For the map grid, J. chose Voronoi diagrams, which symbolize the slippery distances and unusual proportions in dreams (notice that Switzerland, his home land, is much bigger on the map compared to other countries). He used colors to represent emotions (valence and intensity) on the map. Moreover, because for J. feelings and emotions are very important in dreams, he decided to write them down as floating text next to the dream place. Maybe unexpectedly, he represented also flying, as a special ability in dreams, with a simple point symbol on the map at that position, where he could fly. As transportation means present in his dreams there were trains and mountain railways. J. represented cities and localities as well, as they can hardly be omitted on a classical map. In the legend of his map, the difference in size between the real, mapped world and his dream world is also depicted (the dream world is about 10 times bigger).

J. tried to remember all the details in his dreams asked for the setting spiders. On the Web Platform for Dream Cartography (Iosifescu Enescu & Hurni, 2019), he wrote down ten of his dreams and answered all the needed questions, receiving the setting spider (and the event spider as well) in a graphical form. He downloaded these images and, if they had a geographical location (eight out of ten), he placed them on his map (Figure 7).

Figure 7 Personal dreamland with dream setting spiders on it

73 Dream Cartography Concepts

Figure 8 Close-up of personal dreamland, a dream from northern Italy J.’s feedback on the setting metadata profiles was that the spiders made him think about unexpected connections between dream places. Even if dreams took place at different geographical locations, there were other elements that they had in common related to settings, and this was easy to discover by looking at the spider diagrams.

In general, J.’s mapped dreams feature a polygon with a similar area on the setting spider, only pointing in different directions. This means that the dreams have a similar load of unusual elements, but these are different for each dream. In the following, the load on each factor of the setting spider (see Figure 4) is shortly analyzed. The factor to the top, the relation to geographical space, has in most cases a minimum load, so there is nothing unusual: there are geographical locations; the exception is the dream situated in the United States. J. confessed that he had not visited USA at the time of that dream. This setting was a mix between different geographical locations, combined with other sources, such as film material that he watched about similar places. However, in his dream, he was sure that is was USA. Second, the factor society, to the top-right, is as well underrepresented on J.’s setting spiders: the dreams settings were of no different culture, wealth or language, they were practical places, and not very crowded. The third factor, the individual exposure, shows a different pattern. Therefore, the settings near to J.’s hometown, Schaffhausen (northern Switzerland), are clearly familiar, routine and prototypical places, whereas the others are not familiar and no routine places for him, but may possible

74 Paper III

be prototypical places (otherwise the right spider-leg would be fully loaded, as it is the case for the USA and for the Slovenia dreams). The next factor, the attention, seems not to be very unusual: the happenings must have been in line to the setting, the setting received the dreamer’s attention, and they were clearly represented. To the bottom, the role of space is more diverse in these dreams: the place is essential for the happenings (five out of eight dreams), the place could have been replaced without any consequences for the happenings (minimal load, two dreams) and there was one setting which strongly influenced the state of mind (one dream in Schaffhau- sen). Regarding the sixth factor, the morphology, this played a role in these dreams, culminating with the dream in northern Italy (see Figure 8 for a close-up). On a closer look, a violet point- symbol can be observed on J.’s map in this area, which stays for flying in the dream. Therefore the component, which mostly influenced the morphology factor here, is the scale: seeing the setting from a higher perspective, like a bird. To the left, the general exposure is a factor, which was more difficult to bring in the schema usual-unusual (see section 3) and this makes it also more difficult to interpret. In J.’s dreams, this factor mostly plays a role, but never too strong. It can be, that his dreams are in nature, on public spaces, or, e.g., the USA setting was less connected. However, one of the represented dreams has at this factor a minimal load (southern Switzerland), therefore these components were not unusual, or not important for the dream. The last factor, the environment, plays a role in half of the mapped dreams. Because this factor never rises to more than half its load, it can be assumed that in those dreams there were either unusual weather, temperature, lighting conditions, sounds, or physical laws, but not all of them at the same time. Finally, the orientation of the setting spider diagrams to the left or to the right shows where are the most unusual elements, whether in the physical (left), or in the human (right) characteristics of the setting. Most setting spiders profiles are oriented to the right, with two exceptions: one dream in Schaffhausen, where the physical characteristics of the place were more unusual than the social ones; and the USA dream, which looks more like a cross, elongated on the horizontal axis, and with the surface as well slightly wider to the left.

Summarizing, setting spiders were well suited to describe J.’s dream settings. The setting spider is an advance in setting definition and also abstract visualization: as information loaded, multivariate point symbol on the map, when geographical coordinates are available, or just as a stand- alone diagram. If not possible to be placed on a geographical map, the place metadata profiles could be related to each other in terms of time (construction by time), or by relation (e.g. the London tube map, not mainly following geographical relations, but topological ones). Being statistical attributes, the setting spiders could also be showed ordered by any of their attributes, or just by their areas (how unusual the setting is). 6 Summary and Discussion The place cookie and the setting spider models, developed in this work, represent two methods of acknowledging the subjective perception of a setting at a time point. They enable pragmatic comparisons between the subjective perceptions of settings. The place cookie regards only the familiarity of an individual with the place. Because of its simple definition, but also because of

75 Dream Cartography Concepts

the importance of this factor for the perception of a place, its application for categorizing places is straightforward. As a point symbol (five concentric circles) on a map, it offers an instant overview of a person’s knowledge about the depicted locations.

The setting spider model is more complex, being based on a total of 26 properties of a setting. These are the properties, which do make a place distinctive for an individual at a given time. The setting components were assessed with the help of scientific literature, by studying dream reports from an open dream repository and by applying a paper-and-pencil survey. The components were sensibly classified into eight factors, and represented in a radar chart. The left side of the chart comprises factors about the physical properties of the setting, whereas the right side has factors regarding the human-related perception of the setting. The setting spider summarizes the subjective feeling about a setting, on a usual-unusual scale. Represented as a point-symbol for a location on a map, it informs the map reader about the subjective perceived bizarreness of a place during an event. For the sake of completeness, the event or dream spider encapsulate the whole event or dream (dream scene), revealing the topic that was the most important for that event / dream. The event spider model takes as well the form of a radar-chart and can be represented as a multivariate point symbol on a map.

The place cookie is inspired from the social network of a person. Here, the frequency of contact is used for classifying people in circles of acquaintanceship. Similarly, the frequency of interaction with a place can be applied in the case of places. However, it is more difficult to separate places as entities for an individual (scale plays an important role: neighborhood, city, county or country?). To take the surroundings and evaluate them on the familiarity scale is far easier than constructing the whole place cookie with all contained places for a person. Therefore, more research is needed to evaluate how other properties of the social network may be applied to the network of places (e.g. Dunbar’s number).

There are some limitations to the setting spider model as well. First, the question is, if this list of components is indeed exhaustive. For the studied dream reports, the 26 components describe well the subjective perception of the setting. Although not all components are active or perceived in a single dream / dream scene, to the ones that are missing are assigned default values, making these practically “invisible” in the diagram. However, the setting spider was not extensively applied yet. By testing this approach on more participants, its strengths and weaknesses may be revealed. It has to be noticed also that the dream reports (dreambank.net) analyzed for the construction of the spiders come from dreamers living in the USA or western Europe, so there might exist a cultural bias related to the way these dreamers relate to space, respectively to settings, which could be unintentionally reflected in the actual reasoning for the spiders.

The possible correlations between components were considered when these were grouped into factors. Yet there was evidence for the correlations just of a few components (familiarity and routine, exemplified for the survey). Most of the components were assigned to factors through

76 Paper III

theoretical considerations, based on the studied literature, on discussion with experts and on dream examples.

The two maps, where the two proposed methods are applied (Figure 6 and Figure 7), may be classified as univariate, respectively multivariate thematic maps, as they show topographical maps as base maps and load them with additional information. Thematic maps were long used for depicting “phenomena and occurrences of non-topographic type, which however, are related to the earth's surface” (Imhof, 1972, p. 12; Demarmels, Spiess, Schenkel, Heitzler, & Flitter, 2017). The approach for visualizing the characteristics of the setting, proposed in this paper, belongs to thematic cartography. A good comparison can be made between the setting spiders and the pie-charts used for small-scale economic maps in atlases, e.g. in the Swiss World Atlas (Schweizer Weltatlas, 2017). However, dream settings may not always be related to the geographical space, in which case the resulting diagrams cannot be placed on a map; or this space can be deformed in dreams, which is very nicely illustrated by the personal dreamland base map (Figure 7, section 5.2). Nevertheless, the creation of a base dreamland map is not explored in this work, the presented representation being a personal, not reproductible approach.

Moreover, it is suggested that place cookies, setting spiders and event spiders can be used as well for real-life experiences, and not only for dreams. Dream settings are an extreme case of platial representations, but sometimes it is worth going to an extreme, to (maybe) come back with a solution to a more general problem (place versus space in GIS) (A. Zipf, personal communication, September 21st, 2018). Assessing the subjective perception of a real-life place at a time point with the setting spider can be an advantage in many cases, such as in biographical memories (see the second task of the survey, section 3); or in improving the attractiveness of a place based on the type of people visiting the place (e.g. for tourists or for locals). Even if the construction of the setting spiders implies answering 26 single-choice questions, these questions are easy to understand and the advantage of describing the setting in subjective terms is worthwhile. The event spider can be used to visually assess the subjective perception of an event, for example, a public person's rating of an incident in which she was involved; or, on the personal level, as an anchor for a quick overview on events or dreams written in a journal. Yet the application of the proposed methods to real-live events and settings, and therefore their classification in usual or unusual, should be tested in a subsequent work. 7 Bibliography Bernstein, D. M., & Roberts, B. (1995). Assessing dreams through self-report questionnaires: Relations with past research and personality. Dreaming, 5(1), 13. Bollnow, O. F. (1963). Mensch und Raum (Vol. 10). Kohlhammer Stuttgart. Campbell, J. (2008). The hero with a thousand faces. New World Library. (Original work published 1949) Caquard, S., & Cartwright, W. (2014). Narrative cartography: From mapping stories to the narrative of maps and mapping. Taylor & Francis.

77 Dream Cartography Concepts

Couclelis, H., Golledge, R. G., Gale, N., & Tobler, W. (1987). Exploring the anchor-point hypothesis of spatial cognition. Journal of Environmental Psychology, 7(2), 99–122. Demarmels, S., Spiess, E., Schenkel, R., Heitzler, M., & Flitter, H. (2017). Thematic Cartog- raphy. Retrieved from http://www.gitta.info/ThematicCart/en/text/ThematicCart.pdf Dennerlein, K. (2009). Narratologie des Raumes. Walter de Gruyter. Domhoff, G. W. (2003). The scientific study of dreams: Neural networks, cognitive development, and content analysis. Washington, DC: American Psychological Association. Domhoff, G. W. (1996). Finding meaning in dreams a quantitative approach. New York, NY: Plenum Press. Dunbar, R. I. (1993). Coevolution of neocortical size, group size and language in humans. Be- havioral and Brain Sciences, 16(4), 681–694. Dunbar, R. I. (1998). The social brain hypothesis. Evolutionary Anthropology: Issues, News, and Reviews: Issues, News, and Reviews, 6(5), 178–190. Dunbar, R. I. (2010). How Many Friends Does One Person Need?: Dunbar’s Number and Other Evolutionary Quirks. Faber & Faber. Erlacher, D., & Schredl, M. (2004). Dreams reflecting waking sport activities: A comparison of sport and psychology students. International Journal of Sport Psychology, 35(4), 301–308. Falvey, M., Forest, M., Pearpoint, J., & Rosenberg, R. (1997). All my life’s a circle: Using the tools–Circles, MAPS and PATH. ERIC. Foucault, M. (2008). Of other spaces*(1967). In Heterotopia and the City (pp. 25–42). Routledge. (Original work published 1967) Hall, C. S., & Van de Castle, R. L. (1966). The content analysis of dreams. New York: Apple- ton-Century-Crofts. Hauthal, E., & Burghardt, D. (2013). Extraction of Location-Based Emotions from Photo Plat- forms. In J. M. Krisp (Ed.), Progress in Location-Based Services (pp. 3–28). Springer Berlin Heidelberg. Hermann, M. (2010). Visualizing politics and the two kinds of smart people [TEDx Talks]. Retrieved from https://www.youtube.com/watch?v=Uaa07M1f9s0 Hermann, M., & Leuthold, H. (2001). Weltanschauung und ihre soziale Basis im Spiegel eid- genössischer Volksabstimmungen. Swiss Political Science Review, 7(4), 39–63. Hermann, M., & Leuthold, H. (2003). Atlas der politischen Landschaften: ein weltanschauli- ches Porträt der Schweiz. vdf Hochschulverlag AG. Hill, R. A., & Dunbar, R. I. (2003). Social network size in humans. Human Nature, 14(1), 53– 72. Imhof, E. (1972). Thematische Kartographie. Berlin - New York: Walter de Gruyter. Iosifescu Enescu, C. M., & Hurni, L. (2017). Fictional volunteered geographic information in Dream Cartography. International Journal of Cartography, 3(1), 76–87. https://doi.org/10.1080/23729333.2017.1301627

78 Paper III

Iosifescu Enescu, C. M., & Hurni, L. (2019). Cartographic Tools for Mapping Dreams. Pro- ceedings of the 29th International Cartographic Conference. Tokyo, Japan. Iosifescu Enescu, C. M., Montangero, J., & Hurni, L. (2015). Toward Dream Cartography: Mapping Dream Space and Content. Cartographica, (50.4), 224–237. Kligler-Vilenchik, N., Tsfati, Y., & Meyers, O. (2014). Setting the collective memory agenda: Examining mainstream media influence on individuals’ perceptions of the past. Memory Studies, 7(4), 484–499. Lefebvre, H. (1991). The production of space (Vol. 142; D. Nicholson-Smith, Trans.). Oxford Blackwell. (Original work published 1974) Lotman, Y. M. (1977). The structure of the artistic text. University of Michigan Press. (Original work published 1971) Montello, D. R. (1993). Scale and multiple psychologies of space. European Conference on Spatial Information Theory, 312–321. Retrieved from http://link.springer.com/chapter/10.1007/3-540-57207-4_21 Pearpoint, J. (1991). From behind the piano: The building of Judith Snow’s unique circle of friends. Inclusion Press. Reuschel, A.-K., & Hurni, L. (2011). Mapping Literature: Visualisation of Spatial Uncertainty in Fiction. The Cartographic Journal, 48(4), 293–308. Rosen, M., & Sutton, J. (2013). Self-Representation and Perspectives in Dreams. Philosophy Compass, 8(11), 1041–1053. Rosling Rönnlund, A. (2016). Dollar Street - photos as data to kill country stereotypes. Re- trieved February 21, 2019, from https://www.gapminder.org/dollar-street/about Russell, J. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161–1178. Schredl, M. (2003). Continuity between waking and dreaming: A proposal for a mathematical model. Sleep and Hypnosis, 5(1), 26–39. Schweizer Weltatlas (EDK Schweizerische Konferenz der kantonalen Erziehungsdirektoren (Publ.)). (2017). Zurich: Lehrmittelverlag Zürich. Steele, F. (1981). The sense of place. CBI Pub. Co. Strauch, I., & Meier, B. (1996). In search of dreams: Results of experimental dream research. New York, NY: SUNY Press. Tilley, C. Y. (1994). A phenomenology of landscape: places, paths, and monuments (Vol. 10). Berg Oxford. Trechsel, A. H. (2011). EU-Profiler : positioning of the parties in the European elections. Re- trieved from http://bib1rad.iue.private:8080/xmlui/handle/123456789/15 Tuan, Y.-F. (1977). Space and Place: The Perspective of Experience. Minneapolis, MN: U of Minnesota Press. Westerholt, R., Mocnik, F.-B., & Zipf, A. (2018, October). Introduction to the PLATIAL’18 Workshop on Platial Analysis. Retrieved from https://zenodo.org/record/1475267#.XGR22ammkWo

79 Dream Cartography Concepts

80 Paper IV

Chapter 4. Proof-of-Concept Implementation

81 Proof-of-Concept Implementation

Paper IV. Cartographic Tools for Mapping Dreams This is the Author’s Original Manuscript (AOM) of an article to be published in the Proceedings of the 29th International Cartographic Conference in Tokyo, Japan, July 2019

Cristina M. Iosifescu Enescu, Lorenz Hurni

ETH Zurich, Institute of Cartography and Geoinformation, [email protected], [email protected]

Abstract: Creating maps for their dreams enable dreamers to better attend to them. However, mapping dreams is not an easy task due to the particularities of the dream space. Therefore, there is a need of specific cartographic tools for this purpose. This work illustrates the process of creating a Web platform for mapping dreams, functional requirements are stated, and different possible implementations are discussed. By means of a dream example, a dream map and other related visual elements are created, presented and explained. The theoretical framework is based on the Dream Cartography project and the specific diagrammatic visualization tools were developed in its frame.

Keywords: Dream Maps, Dream Cartography, Interactive Web Mapping Platform 1 Introduction Dreams are very personal experiences, which enrich our world. Visual elements such as maps can facilitate a better recalling of a dream, leading also to seeing it from another perspective, eventually understanding more. It has been argued (C. M. Iosifescu Enescu & Hurni, 2017), that dreamers themselves should create maps for their dreams, as they have at their disposal the most complete information. However, creating a dream map is not an easy task. We propose an interactive Web platform, where dreamers can log their dreams and are then provided with specific cartographic tools for creating dream maps.

82 Paper IV

2 Background The research area on dream cartography proposes a view on dreams, where the dream settings or the dream locations play a greater role than they used to in the traditional dream analysis. This is analogous to the spatial turn in humanities, whereas the focus does not shift from time to space (Warf & Arias, 2008), but form the social dimension to the spatial dimension.

Research on dreams has revealed that a great share of the dream elements originate in autobiographical memories, like in the use case presented in section 6; however, a dream rarely replays an awaken experience entirely with all its elements (Malinowski & Horton, 2014). The thought that the brain goes to a (place) memory library before starting to dream is inspiring. Yet in the dreams, there are some places, which are highlighted, such as the childhood home (C. M. Iosifescu Enescu, 2016) or other meaningful places, to which more memories lead.

Moreover, it has been observed that dreams reflect waking-life concerns of people, which is the basis of the continuity theory of dreams (Hall & Nordby, 1972). Some activities (e.g. walking outside, talking with friends) are more probable to appear in dreams (Schredl & Hofmann, 2003), compared to other, cognitive ones (e.g. reading, working on the computer). Whereas the dream cartography project relates more to the cognitive theory of dreams, studying the manifest content of dreams, the traditional preoccupation with dreams in the psychoanalysis of Freud, Jung or Lacan cannot be ignored. A connection point can be seen in the visual reformulation of a dream, performed by the dreamers themselves, or by collecting data about how the dream elements relate to the previous experiences of the dreamers (designating dream elements with their meaning, e.g., Zurich as a familiar place or Mr. Jones as a close friend, a caring person). This implies a kind of “interpretation” of the manifest dream content, leading to a better self- understanding. Moreover, this “abstraction” of real geographical places or person names is similar to using relative coordinates instead of absolute, and makes in the end dreams of different persons comparable to each other by the specific characteristics of the dream elements.

Mapping and logging information about the dream settings could reveal a form of symbolic and archetypal element of the dream language. Dream researchers (Griffith, Miyagi, & Tago, 1958; Nielsen et al., 2003; Schredl, Ciric, Götz, & Wittmann, 2004) have developed and employed a questionnaire of typical dreams to assess how often do some type of dreams occur in different cultures. From a list of 55 dream themes, there are quite a few, which involve a special awareness of the setting, being it the landscape, its morphology, weather phenomena or room arrangement. On this list there are, for example: falling, being on the verge of falling, swimming, flying or floating through the air, floods or tidal waves, tornados or strong winds, earthquakes, fire, school, being at a movie, discovering a new room at home, being unable to find a toilette, losing control of a vehicle, travelling to another planet, being a child again, etc. Researchers applying this questionnaire found differences between men and women or between cultures (e.g., Amer- ican compared to Japanese) in the frequency of these typical dreams. These dream themes are general descriptions, because of the same reason that the dream cartography project turned to

83 Proof-of-Concept Implementation

relative locations, centered on the experience of a person, instead of using (only) geographical locations. The relative description can suit everyone, not only a specific culture or place.

Sharing dreams on social media or on blog sites is a common practice nowadays, but there are no tools yet to visualize dream elements or to systematically assess meaningful statistics related to dream settings. However, visualization could help getting new insights on dreams. Dealing with the visualization and with the additional data needed for it, makes the dreamer be aware of dream elements, which could otherwise get lost. There are benefits of attending to dreams, such as increasing self-understanding, overcoming fears, enhancing creativity or problem-solving strategies (Garfield, 1995).

The target group of the present article comprises cartography scholars, who may be more aware of their surroundings then other people. Therefore, the space, the topography and the atmosphere of a place may play a greater role in their dreams. The results of the research project on dream cartography were implemented in a prototype Web application. In the following, the requirements of such a Web Platform are discussed. 3 Requirements for the Web Platform for Mapping Dreams The workflow for dream cartography consists of data acquisition, data modelling, setting visualization, the visualization of other psychologically relevant dream aspects and the visualization of dream series (C. M. Iosifescu Enescu et al., 2015, p. 227). Based on these topics, the requirements for the Web Platform for mapping dreams are:

a) upload and save dream reports and additional information related to dreams b) create and adjust maps for the dream space, even if this space is mixed or distorted c) visualize dream space, even if it has no geographical reference d) visualize social interactions, emotions present in dreams, and goal-related behavior e) create diagrams for visualizing dream series. f) offer background information about dream research and cartography, or links to such information Besides these technical requirements, for a fully functional Web platform there are also non- functional requirements, such as security, reliability or cost-effectiveness. These will not be specifically handled here.

Point a) from the functional requirements includes a hidden challenge: How to know which additional elements are important to be saved beside the dream report? This requires the modelling of the dream data into clear aspects such as settings, emotions, or social interactions, so that specific questions may be formulated and eventually be answered by the dreamer about each of these aspects.

84 Paper IV

Yet the most challenging requirements refer to the visual elements mentioned in points b) to e). Dream space may show particularities, which make the representation on standard geographical maps (point b) difficult. For the representation of mixed geographical spaces, different methods were pondered, such as location morphing, puzzle, map collage, superimposing of geographic places and cartograms. Morphing in this case could be used to seamlessly join different regions forming the dream space. Cartograms, deforming the geographic space according to whether locations are mentioned in a dream or not, could be used to stress the importance of some locations over others. Moreover, the uncertainty visualization proposed by Reuschel und Hurni (2011) may be used or further adapted for dreams. The authors use, for example, a centered point symbol with fading radial lines extending to the boundaries of the approximate region, where the dream setting could have been located. Another idea is to use customized symbols on the map. These could be sketches created automatically from photographs using, e.g. line jittering with open source software such as GIMP1 or Handy2.

If a dreamed place is not brought in connection with any real place (point c), then the creation by the user of a map from scratch to fit the place description should be possible, using drawing tools. Even the absence of a proper map may be considered. However, in every case, alternative visualization methods such as diagrams or graphs may be created for describing the characteristics of dream space.

Point d) refers to the visualization of elements obtained from the modelling of dream data. Depending on the aspect, the representation may take different forms such as a graph-like-network or a timeline. Specific diagrams such as pie-charts or bars, which incorporate a multitude of aspects, can also be a solution.

For example, the time may be considered in multiple granularities: the time flow inside one dream (the storyline guide) and the timeline for a dream series (point e), which can be differ- entiated in real time and in time related to the number of recorded dreams. Moreover, metadata about time can be recorded and visualized: the story can play in a contracted, expanded or fluent way and it might take place in the present, past or future of the dreamer.

The social interactions in a dream are very important and they may be visualized in a valued directed graph, where the edges represent the value of an interaction (intensity and /or valence) and the nodes are the characters occurring in the dream. The nodes could also encode variables describing the characters (e.g. human / animal, familiar / not familiar).

An important element in dreams is the leading emotion. In the search for a visual representation of emotions, some options are to use for example emoticons, or to change the map style or theme depending on the emotions.

1 https://www.gimp.org/ 2 https://www.gicentre.net/software/#/handy/

85 Proof-of-Concept Implementation

The style of a Web map (color, form, size or orientation of symbols, opacity, etc.) can be adapted for a specific requirement (I. Iosifescu Enescu, 2011). Figure 1 shows an example of predefined styles (e.g., in the middle, the “treasure map” style) applied from the user-interface with Styled Layer Descriptor (SLD) to a set of data provided (from the own geo-server) as a Web Map Service (WMS). The resulting maps contain the same amount of information, yet they convey a very different message regarding their purpose. Whereas high customization can be achieved nowadays applying e.g. Mapbox GL JS style specification3 on vector tiles, more common is the use of pre-rendered (pre-styled) raster tiles. This latter approach is faster from the point of view of the end-user, but it is limited to the available services. Examples are the well-known artistic “Watercolor” or “Toner” from Stamen4.

Figure 1 Predefined styles for GIS data visualization in a Web application (Probst, 2013) Another idea for representing emotions is to use a frame for the map, similar to the way in which, for example, a painting can subtly change its perceived qualities by using a beautiful antique craved wooden frame instead of a modern, simplistic one. Different frames can be designed for basic emotions. Both color and form of the frame can be used to convey the leading emotion. A source of inspiration can be the study of Pixar (2017) for their movie “Inside Out”, where emotions are personified by different characters living inside the head of a child.

Although the dreamers are supposed to create visual elements partly themselves, a text analysis of the dream report is indispensable. In this respect, two methods can be used: asking the dreamers to analyze the dream text themselves or applying external tool for text analysis, such as the machine-learning tool indico.io5 or the text reading and analysis environment from Voyant Tools6.

First, the dreamer should have the possibility to annotate the dream report, choosing words (or word sequences) and classifying these into specific categories: e.g., settings, characters, objects or emotions. These annotations can be used for creating a word cloud. Alternatively, an external tool specialized in visually analyzing text corpus can be used for creating word clouds or other

3 https://docs.mapbox.com/mapbox-gl-js/style-spec/ 4 http://maps.stamen.com 5 https://indico.io/blog/docs/indico-api/text-analysis/ 6 https://voyant-tools.org/docs/#!/guide/about

86 Paper IV

statistical visualization of the text. Figure 2 shows a word could for the text of the dream represented in section 93, generated with Voyant Tools.

Another important contribution of the dreamer is answering additional questions, such as the questionnaires for creating the event spider and the setting spider (C. M. Iosifescu Enescu, Bär, Beilstein, & Hurni, in press), which visualize metadata regarding the dream elements, respectively the dream setting.

Figure 2 Word cloud generated with Voyant Tools Representing dream series (point e) invokes again the different aspects of dreams, which may be quantitatively represented in charts and could offer a different insight in the dreams of a person. For example, the character occurrence for a dream series may be presented in a diagram- like manner, which allows an overview of most frequent character types together with the number of characters occurring in each dream, like in the ThemeRiver, proposed by Havre, Hetzler and Nowell (2000). Moreover, if possible, a trajectory of dream spaces on a map may be shown. A place cookie (defined in (C. M. Iosifescu Enescu et al., in press)) can reveal the distribution of dreams in a dream series based on the familiarity of a dream setting (from 1 very familiar to 6 not known). Figure 3 shows an example of a place cookie for a dream series, featuring also the emotional valence of the dream (positive +, negative – or neutral =).

87 Proof-of-Concept Implementation

Figure 3 Place cookie summarizing the dreams in a dream series based on the dream setting familiarity (1-6) and showing also the emotional valence of the dream (+/-/=) Different (visual) analysis, e.g. on the place cookie, could be performed in different granularities: for a single dream, for a selection of dreams, for a whole dream series, or for a selection of dreams of different dreamers.

Regarding the last functional requirement, point f), important information on dream research and dream cartography can be linked as publications or summarized on the platform. 4 Initial Design of the User Interface In order to create a Web platform for mapping dreams, the functionality was defined (see the previous section with the requirements), then a proof-of-concept design was build, followed by choosing the technologies and a first implementation in form of a functional prototype.

Based on the requirements, a mock-up user interface was created at the beginning of this project (Figure 4). Some of the above listed possibilities were already included, but some came later and were only considered for the prototype phase.

88 Paper IV

Figure 4 Initial proof-of-concept mock-up for the design of the user interface of the Web platform for mapping dreams The largest area is occupied by a geographical map, with two point-symbols on it referring to the geographical locations occurring in the dream. The point symbols are in form of puzzle pieces that fit together, to reflect the fact that, in the represented dream, the two geographic places were actually combined into one place. Additional elements are the filters at the top, allowing to choose a dream from an open dream database, filtering on the dreamer, on the type of the landscape or on types of characters occurring in the dream. To the top right, a small overview map is placed. The dream report and a text with remembered information related to the dream are represented on the right. An orange line in the dream text gives the impression that it may be moved into the dream report and the thought was to represent the places and the social interactions depending on the dream sequence recognized from the text. At the bottom right, a network-graph represents the characters and the social interactions. At the bottom, a big ThemeRiver flow represents the quantitative occurrence of characters in the dream series and under it, a grey line represents the real time with the black lines on it meaning the recorded dreams. This could be navigated with the small black arrows at the ends of the grey line or with the orange line, which also shows the position of the current displayed dream.

89 Proof-of-Concept Implementation

5 A Functional Prototype The prototype phase of the project consists in the implementation of the most important requirements into a functional Web platform.

The first requirement for the Web Platform for mapping dreams is to be able to upload and save dream reports and additional information related to dreams. This is achieved using a Post- greSQL Database on the server side and connecting the HTML and JavaScript form to it with Java servlets and JDBC through an Apache Tomcat Web Server.

Moreover, user registration is introduced, for being able to later edit dream elements, or to create a dream series, but also due to privacy concerns (the non-functional requirement of security was mentioned earlier). The authentication is implemented in Java on the server side using the Blowfish password hashing algorithm (Provos & Mazieres, 1999); the client’s browser communicates with the server over a secure https connection. Users have therefore the possibility to manage their account, to add or delete dreams, or to edit a previously created dream map.

Regarding the text analysis, first, users can use an annotation tool implemented on the platform for classifying words or word sequences from the dream report into four categories: settings, characters, objects and emotions. Second, the machine-learning tool indico.io is used for determining the input language of the dream text. If the dream report is written in English (the analyzing tool is only available for English), then the text is analyzed on place names, person names, emotions and general theme. The emotions are used to suggest a frame in the map application.

The dream setting map is conceived as a do-it-yourself map. The map contains three to four HTML canvas layers, situated on top of each other, which can be reordered by the user. These are one or two geographical map canvases, a frame canvas and a free drawing canvas. The map canvases are powered by the OpenLayers7 library. The advantage of using two map canvases is that it is possible to have different zoom levels, centers and rotations for each canvas, which is useful for creating map collages. The platform allows also to drag-and-drop external vector files containing geographical data (KML, JSON, GPX or IGC) onto the map, if needed.

The frame canvas contains frames such as picture frames, special form frames or meaningful colored frames. The main role of these frames is, beyond the nice finish effect, to convey the emotion of the dream. Different frames were designed, using different colors and forms, for the following basic emotions: sadness, fear, joy, disgust, anger and excitement, based on a Pixar study mentioned in section 3.

In the following lines, the implemented graphical tools, the implemented analytical tools and the graphical elements available for the prototype of the Web mapping platform are listed.

7 https://openlayers.org/

90 Paper IV

Implemented graphical tools related to map creation

§ Draw on map (georeferenced) § Draw on canvas (not georeferenced) § Add own vector geodata § Add own thumbnails § Cut and combine maps (produce map collages) § Select frame § Manage layers and canvases

Implemented analyzing tools

§ Text annotation (user-performed) § Text analyzing with machine learning (external service) § Geocoding (external service) § Questionnaires for detailed profiles § Manage dreams in a dream series

List of representation types on the dream platform

§ Dream map (Do-It-Yourself) § Word cloud § Event Spider Profile § Setting Spider Profile

The event spider (Figure 5) is a visual method for summarizing a dream based on the importance of four factor-pairs occurring in dreams: characters and objects, time and setting, emotion valence and intensity, goal pursuing and fortune. The setting spider (Figure 6) details the factor setting from previously mentioned event spider and is automatically constructed after answering 26 questions. These define the setting in terms of its relation to geospace and the role of space, society and morphology, individual and general exposure, attention and environment. The two spider profiles offer an overview of important dream characteristics and make the comparison or aggregation of dreams form a dream series possible.

91 Proof-of-Concept Implementation

Figure 5 Dream / Event Spider Profile of the selected dream, generated on the Web platform for mapping dreams

Figure 6 Setting Spider Profile of the selected dream, generated on the Web platform for mapping dreams The dreamers are also asked to state the place where they were at the time of the dream, and the given place is geocoded using the Geonames database on cities (points) and Nominatim OpenLayers (points, lines and areas); in the end, the users perform the disambiguation themselves, choosing from the suggested places. The geocoded place may be selected to appear on the map or not.

Not all ideas listed in the previous section(s) are implemented in the current Web platform for mapping dreams, as it is the case usually for a prototype. However, the existence of a Web platform for mapping dreams was successfully tested. In the following, it is illustrated by way

92 Paper IV

of an example dream, respectively by the creation of its map (Figure 7), which innovative features there were integrated into our Web platform 6 Application Example: Use Case The dream to be mapped (see the word cloud of its dream report in German in Figure 2) takes place in Zurich, Switzerland, but a part of the dream city is actually Goettingen, Germany. The dreamer (von Uslar, 2003) lived at the point of the dream in Zurich and at an earlier time point in Goettingen. The general dream tone is rather sad, and the dreamer talked to one other person in the dream. This dream has a geolocation; however, this is a mix between two real places. Moreover, the dream has an emotion, characters and social interactions. All these dream elements have to be represented on the dream map.

Figure 7 Example of a dream map created with the interactive Web platform for mapping dreams The dream report is uploaded into the platform like in a dream diary. The text of the dream is automatically analyzed using the external machine-learning tool (indico.io, see section 3), which recognizes the leading emotion and the place names. The system suggests for this dream a rounded blue frame for the map, which stays for sadness. In addition, the user has the possibility to annotate the dream text and to create a word cloud.

Place names in the dream text are geocoded; in the end, the user performs the disambiguation herself, choosing from the suggested places (e.g. Zurich, Switzerland and not Zurich, Kansas, USA).

After the text processing, the user is presented with a map stub. On this map, the places chosen in the previous step, Zurich Switzerland and Goettingen Germany, are marked.

93 Proof-of-Concept Implementation

Setting the map canvas active, the user drags from the initial list with available layers the desired layers to compose her map: the WMS layers Stamen Toner and OpenStreetMap. The lo- cally defined vector layers “geojsonplaces” (contains the markers for the two cities) and “play- ground” (for user-defined features) are already added to the map. Using the map features tools, the user draws a new feature on the map: an area containing the city center of Goettingen; then she clips this area out of the OpenStreetMap WMS layer. Adding a new map canvas from the map layers tool, the user moves the newly resulted layer to the new map canvas. Now she zooms the first map canvas to Zurich and, making the second map canvas active, she then pans the clipped city-center of Goettingen so that it covers a part of Zurich on the first map canvas.

Setting the free drawing canvas active, the user drags and drops two images as thumbnails on the map: two profile pictures from her local computer, representing the characters present in the dream. On this canvas, it is possible to draw like in a paintbrush program, choosing different colors or brush size and types, in order to personalize the dream map as desired. In this case, the user draws an arrow between the two profiles, representing the one-way social interaction. With this, all relevant dream elements are visualized on the map (see Figure 7): it is a sad dream, with a one-way social interaction between two persons, taking place at a geographically mixed location. The textual and visual elements are saved into the database on the server. Each dream can be visualized or edited later, and all dreams uploaded by the same user are saved into a dream series.

As additional visualization diagrams, the user chooses to create the dream spider (Figure 5) and the setting spider (Figure 6), by answering the two specific questionnaires about the dream, respectively dream setting.

In this section it was illustrated that the prototype of the Web platform for mapping dreams presents already many of the desired functionalities. 7 Conclusions and Outlook Proposing a tool for mapping dreams encourages the registration of its geographical aspects. The cartographic tools for mapping dreams could be used also for mapping narratives. Both mapping dreams and mapping narratives present the challenge of mapping fictional worlds. However, a dream is an experience lived by the dreamer, and the dream report is much shorter than a piece of narrative. The latter one is more elaborated and might include more elements to be used for mapping. However, usually a literary expert performs the mapping and has only access to the narrative itself and to external sources of information related to it. On the other side, for a dream, the most interested client is the dreamer herself, and the dreamer might be willing to answer questions related to the dream, preferably shortly after the dream experience. Through specific questions, e.g. the dream setting can be well described by the setting spider. The dreamer is the incontestable authority on the subject, which is different for narratives, if

94 Paper IV

not the authors themselves are performing the mapping. Reflection on dream elements and in- trospection are methods applied by the dreamer in order to understand a dream and mapping one’s dreams foster these by offering tools for visualization and organization.

Recalling the discussion in the beginning, which states that relative descriptions of places are better suited for dream places, it could be nevertheless interesting to relate a dream to a geolocation. This may be possible in the Web platform for mapping dreams. For example, all dreams of everyone who ever dreamed of the Tour Eiffel in Paris, France could be listed and statistics could be displayed on who these people are and what was their real-life experience in this location.

With the help of such a Web platform for mapping dreams, beyond mapping dreams, it may be revealed, for dream a series, which are the most meaningful places, to which the memories and dreams of a person lead, and what emotions are usually experienced in the dreams that take place in these locations. 8 References Garfield, P. (1995). Creative Dreaming: Plan And Control Your Dreams To Develop Creativity Overcome Fears Solve Proble. Simon and Schuster. Griffith, R. M., Miyagi, O., & Tago, A. (1958). The universality of typical dreams: Japanese vs. Americans. American Anthropologist, 60(6), 1173–1179. Hall, C. S., & Nordby, V. J. (1972). The individual and his dreams. New York: New American Library. Havre, S., Hetzler, B., & Nowell, L. (2000). ThemeRiver: Visualizing theme changes over time. Information Visualization, 2000. InfoVis 2000. IEEE Symposium On, 115–123. Iosifescu Enescu, C. M. (2016). Impact of Migration on the Dream Setting (Master’s thesis for the degree of Master of Science in Psychology UZH). University of Zurich, Zurich, Switzerland. Iosifescu Enescu, C. M., Bär, H. R., Beilstein, M., & Hurni, L. (in press, 2019). Place Cookies and Setting Spiders in Dream Cartography. Transactions in GIS. DOI:10.1111/tgis.12604. Iosifescu Enescu, C. M., & Hurni, L. (2017). Fictional volunteered geographic information in Dream Cartography. International Journal of Cartography, 3(1), 76–87. https://doi.org/10.1080/23729333.2017.1301627 Iosifescu Enescu, C. M., Montangero, J., & Hurni, L. (2015). Toward Dream Cartography: Mapping Dream Space and Content. Cartographica, (50.4), 224–237. Iosifescu Enescu, I. (2011). Cartographic Web Services (Dissertation for the degree of Doctor of Sciences). ETH Zurich. Malinowski, J. E., & Horton, C. L. (2014). Memory sources of dreams: the incorporation of autobiographical rather than episodic experiences. Journal of Sleep Research, 23(4), 441–447.

95 Proof-of-Concept Implementation

Nielsen, T. A., Zadra, A. L., Simard, V., Saucier, S., Stenstrom, P., Smith, C., & Kuiken, D. (2003). The typical dreams of Canadian university students. Dreaming, 13(4), 211–235. Pixar. (2017). The art of storytelling, 4. Visual language, Shape (video). Retrieved March 13, 2019, from Khan Academy website: https://www.khanacademy.org/partner-content/pixar Probst, T. (2013). Benutzerdefinierte Symbolisierung von OpenStreetMap-Daten mit QGIS- Server [Interdisciplinary Project]. Retrieved from ETH Zurich, Institute of Cartography and Geoinformation website: http://www.ika.ethz.ch/studium/masterpro- jektarbeit/2013_probst_bericht.pdf Provos, N., & Mazieres, D. (1999). A Future-Adaptable Password Scheme. USENIX Annual Technical Conference, FREENIX Track, 81–91. Reuschel, A.-K., & Hurni, L. (2011). Mapping Literature: Visualisation of Spatial Uncertainty in Fiction. The Cartographic Journal, 48(4), 293–308. Schredl, M., Ciric, P., Götz, S., & Wittmann, L. (2004). Typical dreams: stability and gender differences. The Journal of Psychology, 138(6), 485–494. Schredl, M., & Hofmann, F. (2003). Continuity between waking activities and dream activities. Consciousness and Cognition, 12(2), 298–308. von Uslar, D. (2003). Tagebuch des Unbewussten : Abenteuer im Reich der Träume. Würzburg: Königshausen & Neumann. Warf, B., & Arias, S. (2008). The Spatial Turn: Interdisciplinary Perspectives. Routledge.

Chapter 5. Concluding Remarks

97 Concluding Remarks

Synopsis of Results This thesis introduces Dream Cartography, a new field of research, proposing cartographic methods for the analysis and visualization of dream data. The existing data sources and the acquisition of spatially related data in dreams are discussed, a data model is developed taking into account the subjective and dynamic experience of a place, and possible visual representations for the dream setting and for other psychological aspects of dreams are illustrated. Overall, the research questions formulated at the beginning of this thesis were answered, and the following paragraphs summarize the results.

Dream data sources and improving data acquisition (RQ1) This research question is mainly addressed in the first two publications. Dream reports are com- monly used for analysis in dream research. The careful preparation of the dream recording in advance, the selection of an unstressful day for recording, and dream rehearsal for a better recollection of the dream experience (see Table 1, p. 24) are ways of improving the assessment of a dream report. In addition, answering a suggested questionnaire regarding the dream setting and also other aspects of the dream content (see Table 2, p. 25) can help dreamers to focus and better remember details about their dreams (paper I). Other valuable, complementary data source for dream content are the autobiographical memories of the dreamers. A fictional VGI is proposed, arguing that every dream can be the object of analysis of different volunteers (paper II). The geographical world is recognized as a great source of inspiration for creating new geographies for fictional space, for example with reversed properties, such as north to south (e.g. Figure 3, p. 47).

In paper III, after creating two data models for dream settings, the data acquisition is further refined with specific questionnaires. For the first model, a single question is asked, regarding the place familiarity. For the second, more complex model of the setting, a set of 26 single- choice straightforward questions are asked (Appendix B. ). Additionally, a questionnaire with 8 questions is suggested, regarding the dream content in general (Appendix C. ). The answers to these questionnaires considerably improve the data, which can be used for dream analysis.

Finally, the setting up of a Web platform for mapping dreams is explored (paper IV), featuring the input of text (dream reports) and the possibility to answer all the mentioned questionnaires. Summarizing, the data sources for dream analysis are the dream reports, answers to specifically developed questions, and the biography of the dreamer. Data acquisition can be improved by following the proposed instructions for dream recording and by the development of the needed questionnaires.

Data model of the dream setting (RQ2) Modelling the dream setting is a complex issue, and different elements contributing to the setting are analyzed. Different elements help dreamers to recognize a place in their dreams: specific landmarks, environmental elements or present dream characters. However, from an online survey it turns out that many dreamers just have the feeling to be in a specific place and cannot always express why. Another studied element of the setting is the scale: dreams mostly take place on a large-scale level of space, being experienced from a perspective similar to reality (paper II). The third paper contributes at most to establishing a data model for the dream settings. A list of questions related to space, place and atmosphere in dreams is created based on consulted literature on dream psychology, geography and narratology. These questions are applied to a set of dreams from a public dream repository and the list is updated accordingly. A part of the questions are validated through a survey. Finally, a total of 26 questions remain and are translated into 26 dimensions (see Table 3, p. 67 and Appendix B. ), which extensively characterize a setting. Furthermore, these dimensions are grouped into eight factors. Two of the most relevant dimensions compose directly two factors: the relation of the setting to a geographical location; and the influence of the space on the happenings, stating if the scenery becomes a protagonist or not. The remaining six factors characterize the setting either related to human or physical geography, and form opposite pairs: society vs. morphology, individual vs. general exposure, and attention vs. environment. For a better understanding of this model, the factors are visualized into a profile, named setting spider. The (single-choice) answers to the 26 questions are assigned to a higher or lower probability of occurrence (usual to unusual) and the dimensions contribute with different weights to the composition of a factor. The calculated values of the factors, based on the answers given by a dreamer for a specific dream, are visualized on the setting spider.

It is argued that this comprehensive data model can be also used for real settings of real events, because the identified characteristics of dream settings are relevant as well for real settings. For example, a person’s familiarity with a place is a key variable for its perception both in dreams and in reality. The familiarity of the dreamer with a place is used for building another model of the dream setting, which is the base for the place cookie diagram. Finally, the spiders and the place cookies can be applied as multi-, respectively univariate point symbols on a map, if the geographical location of the setting is known.

99 Concluding Remarks

Visualization of the dream setting on a map (RQ3) A dream setting can realistically portray the geographical space, but can be also a mix between different locations, or be totally invented. Therefore, the basic cartographic visualization of a dream setting is challenging, and this issue is touched in different forms by all papers related to this thesis. The use of different visualization methods is proposed, depending on this relation of the dream space to the geospace (paper I). For imported, geographical spaces, a simple point symbol, route or area on a geographical map can be used, because the dream location is thereby clearly depicted. For geographical but uncertain dream settings, known visualization methods such as fuzzy shapes or textures can be used. In case of mixed, condensed dream settings, a morphing or collage of geographical maps is proposed (Figure 1 and Figure 3 in paper I; Figure 2 in paper II and Figure 7 in paper IV); using puzzle pieces on a map is also possible (Figure 4 in paper IV). A personal dreamland map created by a dreamer is shown in Figure 7 and Figure 8 of paper III. The dreamer chose to represent his dreamland in form of a cartogram. Many geographical places are missing on this map, because the dreamer never dreamed about these places. Although in the first paper a constructed 3D model of a dream setting is shown (Figure 4, p. 33), the following research focuses on abstracting the characteristics of dream settings. For the visualization of places, which have no relation to the geographical world whatsoever, the univariate, respectively multivariate point symbols representing the place cookie or the setting spider (e.g. Figure 6 of paper IV), obtained from data modeling, can be depicted. Alterna- tively, the user of the Web platform for mapping dreams can employ the canvas-drawing tool to sketch the missing spatial relations.

Visualization of dream content (RQ4) Beyond the spatial extension of a dream, there are different important topics in dreams, and their visualization fosters the dream analysis. Therefore, the first paper proposes the depiction of characters in a dream as stylized profiles (see Figure 6 of paper I, p. 36) with features revealing the age group of the persons and their familiarity to the dreamer. To illustrate the social interactions, the characters can be arranged in a network graph, where the edges represent the communication type during the dream (see Figure 5 of paper I). Furthermore, the use of a customized border is proposed, such as a photo frame, for the dream map, which suggests the leading emotion in the dream. Specific frames for basic emotions are prepared, making use of different forms and colors (paper IV). The visualization of character occurrence over a series of dreams can be achieved using a riverflow diagram (Figure 4, bottom, p. 89). The time flow can be illustrated by a timeline both over a series of dreams and in a single dream. Moreover, a data model for dream scenes is proposed (paper III), which comprises eight factors: actors, objects, emotional valence, emotional arousal, goal (active participation), good fortune or misfortune (passive), time and setting. By answering eight single-choice questions regarding these factors, an event spider (see Figure 5 of paper III and Figure 5 of paper IV) is constructed and

100

can be visualized for each dream scene. The values, which are assigned to the answers, estimate the importance of a factor for the given dream (see Appendix C. ). The event spider is applied for a case study in paper IV.

Analysis and comparison of dream settings (RQ5) The analysis and comparison of dream settings is performed based on the developed data models. Different place cookie diagrams, showing the dreamer’s familiarity to a dream place, can be rendered for each dream setting. These can be compared parametrically (value of the given answer) or visually (e.g. how close to the center is the black circle, see Figure 6 of paper III). In addition, different dream settings can be summarized over this characteristic on a single place cookie, depicting the distribution of familiar and unfamiliar places in a series of dreams (Figure 3 of paper IV, p. 88). The setting spider model, featuring the numerous dimensions of a dream setting in a single diagram, is well suited for both parametric and visual comparison of settings in each of the eight factors. The bizarreness of a dream setting can be assessed from the area of the spider (obtained by connecting the points for the respective values on each spider leg). This is due to the common scale applied to the factors by definition: a greater value of a factor means a less usual state. A qualitative, detailed analysis of a setting is made possible by pointing to and discussing the obtained values for the factors of the setting spider, mentioning their components. Similar to the place cookies, a visual comparison of settings can be performed by using the spiders of different dream settings on a single dream map, as multivariate point symbols (see Figure 7 of paper III, p. 73).

Discussion Researching the subject of dream cartography was a challenging task. The setting spider model is one of the most important achievements, because it allows a visual and intuitive comparison of complex platial characteristics. Especially noteworthy is the finding, that all conceivable characteristics of a setting can be used for its modeling and that these characteristics can be grouped into a definite number of meaningful factors. The greatest advantage is that these characteristics are modeled on a common scale (from usual to unusual) and therefore can be visualized in a compact system. However, having the same scale can also be a danger. It is rather easy to estimate how usual or unusual it is to dream about familiar compared to unfamiliar places, or to hear a known language in dreams (compare to the continuity hypothesis of dreams). Yet, estimating how usual is the appearance of a public versus a private space in dreams is rather awkward, and can vary with the previous experiences of dreamers (subjectivity). A sys- tematical study of these variables for a greater number of dreams from different individuals would bring more clearness and readiness to accept these values. The second model, the place cookie, is, on the other hand, not only clear and straightforward, but also very powerful. The

101 Concluding Remarks

analysis, which can be done with this model, may reveal important aspects of the encounter of people with their dream places.

The enhancement of data acquisition by a questionnaire for dream settings results in having available data for the visualization of the setting spider model. However, it does only partially resolve the problem of mapping individual dreams. The solution, to empower the users by F- VGI, is innovative. However, it could also act as a limitation, due to individual differences in understanding a dream report (compare to the previously mentioned interrater reliability) or mapping skills. Proposing the dreamers themselves as the volunteers for mapping their own dreams resolves the problem of interpretation, but not the one of mapping, even if the mapping tools are at disposal.

The proposed mapping tools are useful, but rather limited, compared to the expectations rose with the 3D model presented in the first publication, and to what the word “dreamworld” evo- cate to most people. The dreamland map, on which the setting spiders are used as a thematic layer (paper III), is artistic and repairs a bit the expectations. However, recognizing the irrepro- ducibility of this dreamland map can be disappointing. At least some subtle, interesting details (such as having Voronoi diagrams as grid) are mentioned, which were used for the creation of this map.

Finally, the results of this work were published exclusively in journals and conference proceedings addressing experts in cartography and geosciences. The publication of use cases (e.g. a thematic dream series) or of other research based on this work in journals dedicated to psychology scholars would have been an advantage.

Outlook The field of Dream Cartography is just emerging. Further investigations on the formation of the spatial extent of dreams could lead to a reproducible representation of a base map for a dream or for a dream world. What are the principles to create a map of a whole dream world, such as the one created by J. in paper III, and to continuously update it with every dream? Can automatic morphing of geographical locations be implemented? Research on different geographies proposes also the so-called map-folding (Bergmann & O’Sullivan, 2017), which could be applied to dream locations. Here, the 2D map is transformed in 3D after digitally folding the 2D map like an origami, to connect far-away locations. The use of cartograms for mapping the dream space could also be proved.

The approach to use setting spiders, i.e. the characteristics of dream settings, as multivariate, geotagged point symbols on a geographical map, risks to be too abstract and scientific, thus causing the dream world to lose expressivity and emotional attachment. The opposite would be to reconstruct the dream world for Augmented Reality or for Virtual Reality, which could be a

102

good application in the gaming industry, for example, to create personally relevant gaming en- vironments.

While the developed setting model was tested for dreams, it was not yet tested for real-life settings. The application of such a dynamical model of the setting could bring new insights for touristic places, or for spatial and settlement development in urban environment. It can be assessed not only how a setting changes in time with regard to the environmental conditions or general exposure, but also how different people in the area perceive this place with respect to their previous experiences. Alternatively, the focus could lie on a person, instead of a place. For example, should an individual rate his current whereabouts on the setting spider regularly during the day, would this increase his awareness of the surroundings, increase his mindfulness? Would a systematic classification of places familiar to an individual reveal numbers similar to the concentric circles in the social network (5-15-50-150, see paper III, p. 58), such as by Dun- bar (2010)?

Moreover, an extensive survey to test the developed data models of place cookie, setting spider and event spider could be set up. This could be also done by establishing a Web platform, similar to the one mentioned in paper IV. A Web platform for mapping dreams could foster the use of maps in dream research, promoting cartography as a public tool for organizing, visualizing and gaining knowledge from dreams.

The setting model developed in the frame of the Dream Cartography project could be also applied or adapted to other different geographies, such as literature or narratology. Taking a fairy tale as example, it could be studied how its setting or event spiders evolve throughout the course of the story, e.g. from known to unknown places (Campbell, 1949/2008). Furthermore, it would be interesting to observe, how this succession of spiders differs compared to other literary or non-literary texts. Also, the meaningful (automatic) partitioning of a dream report into dream scenes or the creation of a story board for a dream could be explored, as dreams were since the beginning of time used as ideas for stories, art creations, or lately as film ideas, or for scientific discoveries.

103 References

References References for chapters 1, 2, and 5. For references mentioned in chapters 3 and 4 see the references lists at the end of each paper.

Adaman-Tremblay, L. (2015). My Night Life. Retrieved from http://www.lulu.com/shop/lidia- adaman-tremblay/my-night-life/paperback/product-21982102.html Baylor, G. W., & Cavallero, C. (2001). Memory sources associated with REM and NREM dream reports throughout the night: A new look at the data. Sleep: Journal of Sleep Research & Sleep Medicine, 24, 165–170. Baynton-Williams, A. (2015). The Curious Map Book. University of Chicago Press. Bergmann, L. R., & O’Sullivan, D. (2017). Computing with many spaces: Generalizing projec- tions for the digital geohumanities and GIScience. Proceedings of the 1st ACM SIG- SPATIAL Workshop on Geospatial Humanities, 31–38. ACM. Bernstein, D. M., & Roberts, B. (1995). Assessing dreams through self-report questionnaires: Relations with past research and personality. Dreaming, 5(1), 13. Bollnow, O. F. (1963). Mensch und Raum (Vol. 10). Kohlhammer Stuttgart. Campbell, J. (2008). The hero with a thousand faces. New World Library. (Original work published 1949) Cartographies of Fictional Worlds. (2011). The Cartographic Journal, 48(4), 218–223. https://doi.org/10.1179/174327711X13190991350051 Cipolli, C., & Poli, D. (1992). Story structure in verbal reports of mental sleep experience after awakening in REM sleep. Sleep: Journal of Sleep Research & Sleep Medicine. Cresswell, T. (2014). Place: an introduction. John Wiley & Sons. DeGraff, A., & Harmon, D. (2015). Plotted: A Literary Atlas. Pulp. DeGraff, A., & Jameson, A. D. (2017). Cinemaps: An atlas of 35 great movies. Quirk Books. Desseilles, M., Dang-Vu, T. T., Sterpenich, V., & Schwartz, S. (2011). Cognitive and emotional processes during dreaming: A neuroimaging view. Consciousness and Cognition, 20(4), 998–1008. Domhoff, G. W. (1996). Finding meaning in dreams a quantitative approach. New York, NY: Plenum Press. Dunbar, R. I. (2010). How Many Friends Does One Person Need?: Dunbar’s Number and Other Evolutionary Quirks. Faber & Faber. Fosse, M. J., Fosse, R., Hobson, J. A., & Stickgold, R. J. (2003). Dreaming and episodic memory: a functional dissociation? Journal of Cognitive Neuroscience, 15(1), 1–9. Foucault, M. (2008). Of other spaces*(1967). In Heterotopia and the City (pp. 25–42). Routledge. (Original work published 1967) Foulkes, D. (1999). Children’s dreaming and the development of consciousness. Harvard Uni- versity Press. Freud, S. (1942). [1900] Die Traumdeutung (A. Freud, Ed.). London: Imago.

104

Hall, C. S., & Van de Castle, R. L. (1966). The content analysis of dreams. New York: Apple- ton-Century-Crofts. Harmon, K., & Harmon, K. A. (2004). You are here: Personal geographies and other maps of the imagination. Princeton Architectural Press. Hau, S. (2006). Traumforschung. In H. Böker (Ed.), Psychoanalyse und Psychiatrie (pp. 359– 369). Heidelberg: Springer Medizin Verlag. Horikawa, T., Tamaki, M., Miyawaki, Y., & Kamitani, Y. (2013). Neural Decoding of Visual Imagery During Sleep. Science, 340(6132), 639–642. Hurni, L. (2015). Kartographie: Visualisierung von Objekten und Phänomenen in Raum und Zeit. Ein Essay in zehn Aspekten. In Welt-Anschauungen: interdisziplinäre Perspekti- ven auf die Ordnungen des Globalen (Vol. 67, pp. 55–78). Deutsche Akademie der Na- turforscher Leopoldina. Iosifescu Enescu, C., Iosifescu Enescu, I., & Hurni, L. (2013). Customizable Map Content and 3D Visualization in the Online Platform GeoVITe. Proceedings of the 26th Interna- tional Cartographic Conference. Presented at the 26th International Cartographic Con- ference, Dresden, Germany. Kriz, K., Cartwright, W., & Hurni, L. (2010). Mapping Different Geographies. Berlin: Springer. Lefebvre, H. (1991). The production of space (Vol. 142; D. Nicholson-Smith, Trans.). Oxford Blackwell. (Original work published 1974) Lewis-Jones, H. (Ed.). (2018). The Writer’s Map: An Atlas of Imaginary Lands. Thames & Hudson Ltd. Montangero, J. (2012a). Cognitive Approach to Dreaming. In D. Barrett & P. M. Ph.D (Eds.), Encyclopedia of Sleep and Dreams (pp. 133–135). Santa Barbara, California: ABC- CLIO. Montangero, J. (2012b). Dreams are narrative simulations of autobiographical episodes, not stories or scripts: A review. Dreaming, 22(3), 157–172. Montello, D. R. (1993). Scale and multiple psychologies of space. European Conference on Spatial Information Theory, 312–321. Retrieved from http://link.springer.com/chapter/10.1007/3-540-57207-4_21 Moretti, F. (1999). Atlas of the European novel, 1800-1900. Verso. Peterson, M. P. (2008). International perspectives on maps and the Internet. Berlin Heidelberg New York: Springer. Piatti, B. (2008). Die Geographie der Literatur. Schauplätze, Handlungsräume, Raumphanta- sien, 2. Piatti, B., Bär, H. R., Reuschel, A.-K., Hurni, L., & Cartwright, W. (2009). Mapping literature: towards a geography of fiction. In W. Cartwright, G. Gartner, & A. Lehn (Eds.), Car- tography and Art (pp. 177–192). Berlin Heidelberg: Springer. Piatti, B., Reuschel, A.-K., & Hurni, L. (2011). A Literary Atlas of Europe-Analysing the Ge- ography of Fiction with an Interactive Mapping and Visualisation System. Proceedings of the 25th International Cartographic Conference, 3–8.

105 References

Reuschel, A.-K. (2012). A Literary Atlas of Europe (Poster) : Ein Literarischer Atlas Europas. Retrieved March 5, 2013, from http://www.literaturatlas.eu/files/2012/03/Literary_At- las_of_Europe_POSTER_gut.pdf Reuschel, A.-K., & Hurni, L. (2011). Mapping Literature: Visualisation of Spatial Uncertainty in Fiction. The Cartographic Journal, 48(4), 293–308. Reuschel, A.-K., Piatti, B., & Hurni, L. (2013). Modelling Uncertain Geodata for the Literary Atlas of Europe. In K. Kriz, W. Cartwright, & M. Kinberger (Eds.), Understanding Different Geographies (pp. 135–157). Berlin Heidelberg: Springer. Ribeiro, D. M., & Caquard, S. (2018). Cartography and Art. The Geographic Information Sci- ence & Technology Body of Knowledge. https://doi.org/10.22224/gistbok/2018.1.4 Rosling Rönnlund, A. (2016). Dollar Street - photos as data to kill country stereotypes. Re- trieved February 21, 2019, from https://www.gapminder.org/dollar-street/about Sandler, L., Kramer, M., Fishbein, H., & Trinder, J. (1969). Interlaboratory reliability of Hall- Van de Castle scale. Psychophysiology, 6, 248–249. NEW YORK, NY: CAMBRIDGE UNIV PRESS. Schredl, M. (2010). Characteristics and contents of dreams. In International review of neurobi- ology (Vol. 92, pp. 135–154). Elsevier. Schredl, M., Burchert, N., & Gabatin, Y. (2004). The effect of Training on Interrater Reliability in Dream Content Analysis. Sleep and Hypnosis, 6, 139–144. Steele, F. (1981). The sense of place. CBI Pub. Co. Strauch, I., & Meier, B. (1996). In search of dreams: Results of experimental dream research. New York, NY: SUNY Press. Tilley, C. Y. (1994). A phenomenology of landscape: places, paths, and monuments (Vol. 10). Berg Oxford. Tuan, Y.-F. (1977). Space and Place: The Perspective of Experience. Minneapolis, MN: U of Minnesota Press. Vaughan, L. (2011). Mapping the imagined. In Mapping different geographies (pp. 93–102). Springer. von Uslar, D. (2003). Tagebuch des Unbewussten : Abenteuer im Reich der Träume. Würzburg: Königshausen & Neumann.

106

Appendices

107 Appendices

Appendix A. Survey for setting profile Dream-scene setting profile (December 2018)

I Please think about a dream you remember vividly. This dream had one or several settings. Put an A to each of the statements, to each you agree.

II Please recall a vivid biographical memory, preferably something from your childhood. Put an X to each of the statements, to each you agree.

To me, this setting... 1 Familiarity 100 Was familiar 0 Was not for sure familiar or not -100 Was not familiar

2 Routine 100 Is or used to be a routine place (work, school) 0 Was not for sure a routine or no routine place -100 Is not and never was a routine place

3 Time 100 Was in the current time 50 Was in my past 0 Was in the near future -50 Was in historic times, or in the far future -100 Had no defined time

4 Relation to geospace 100 is geographical 50 is a mix between different geographical locations 0 is a condensed setting -50 is an invented setting at a known geographical location -100 is an imagined setting

5 Function of space 100 strongly influenced (physical or psychological) the dream 0 was essential for the happenings -100 is replaceable without any consequences to the dream

6 Accessibility 100 Was a public space 0 Was not for sure a public or private space -100 Was a private space

7 Spiritual / Practical 100 Was a spiritual or magical space 0 Was neither magical nor useful space, or both -100 Was a practical or useful space

108

8 Congruence, Expec- 100 Was in line with the happenings tancy 0 Was neither in line nor in contradiction with the happenings -100 Was in contradiction with the happenings

9 Attention 100 Comprised an active moment (contemplation, bellevue point) 0 Received my attention -100 Was in the background

Elicited or was related to a pleasant emotion (safety, joy, 10 Atmosphere / Mood 100 etc.) 0 Elicited both pleasant and unpleasant emotions -100 Elicited an unpleasant emotion (fear, sadness, anger, etc.).

11 Crowded 100 Was very crowded 50 was crowded 0 was agreeably uncrowded -50 had only a few people around -100 was deserted

12 Space Volume 100 Was spacious and vast 0 was neither spacious nor restrictive -100 was restrictive and confining

13 Light 100 was bright 0 was not clearly illuminated -100 was dark

14 Temperature 100 was very hot 50 was warm 0 had a nice temperature -50 was cold -100 was very cold

15 Complexity 100 was very complex and detailed 0 had a moderate complexity -100 had a low complexity

16 Clarity of represen- 100 was clearly represented tation 50 was partly clearly represented 0 was blurred and had no details -50 was known but not represented -100 was indistinct

109 Appendices

17 Scale 100 was viewed from the normal perspective 0 was viewed from a higher perspective (bird) -100 was viewed from a lower perspective (ant)

18? ------

110

111 Appendices

Appendix B. Statements and their values for the Setting Profile Setting spider components with their exact statements and corresponding values (0 = usual, 100 = unusual), single choice; detail of Table 3 from Paper III, Chapter 3

To you, this (dream) setting…

Component Value Description

0 is geographical 33 is a mix between different geographical locations Relation to Geospace 66 is an invented setting at a known geographical location 100 is an imagined setting at an unknown geographical location 0 had only a few people around 50 was agreeably crowded Population Density 100 was very crowded 100 was deserted 0 had more or less the same wealth level I'm used to Wealth 50 was richer than I am used to 100 was poorer than I am used to 0 mine / not perceived Language 50 known 100 unknown 0 was the same culture as mine/ or not perceived 25 was a known, similar culture to mine Culture 50 was a known culture, but different than mine 75 was an unknown culture, but similar to mine 100 was an unknown, different culture 0 was a practical or useful space Spiritual/Practical 50 not sure / was neither magical nor useful space, or both 100 was a spiritual or magical space 0 was familiar Familiarity 50 not sure / was both familiar and unfamiliar 100 was not familiar 0 is or used to be a routine place Routine (e.g. work, school) 50 not sure / was not for sure a routine or no routine place 100 is not and never was a routine place Prototypical Place, categoriz- 0 was a prototypical place able (e.g. restaurant, war, 50 not sure Western, Sci-Fi) 100 was not a prototypical place Attention 0 received my attention

112

50 was in the background 100 comprised an active moment (contemplation, bellevue point) 0 was blurred and had no details 50 was partly clearly represented Clarity of Representation 50 was known but not represented 100 was clearly represented 100 was indistinct 0 was in line with the happenings 50 not sure / was neither in line nor in contradiction with the Congruence happenings 100 Was in contradiction with the happenings 0 is replaceable without any consequences to the dream 33 was essential for the happenings Role of Space 66 strongly influenced the state of mind (threatening, free- dom) 100 took physical action (natural hazards) 0 was viewed from the normal perspective Scale 50 was viewed from a higher perspective (bird) 100 was viewed from a lower perspective (ant) 0 not sure / was neither spacious nor restrictive Space Volume 50 was restrictive and confining 100 was spacious and vast 0 had no altitude differences / not perceived Landscape/Curvature 50 had moderate altitude differences 100 had high altitude differences 0 had a low complexity Complexity 50 had a moderate complexity 100 was very complex and detailed 0 not perceived 50 was moderately neat Neatness 100 was very neat 100 was very messy 0 was a private space Accessibility 50 was a public space 100 not sure / was not for sure a public or private space 0 not sure 25 was physically and virtually well connected Connectivity 50 was physically well connected 75 was virtually well connected 100 was not connected 0 had both artificial and natural elements Natural/Artificial 50 was predominantly man-made/artificial 100 was predominantly natural

113 Appendices

0 physical laws were respected Physical Laws (e.g. Gravity) 100 were only partially respected / were different 0 had a nice weather / not perceived 50 was affected by middle weather conditions (precipitations, Weather wind) 100 was affected by strong weather conditions (tornado, fire) 0 had a nice temperature / not perceived 50 was warm Temperature 50 was cold 100 was very hot 100 was very cold 0 was bright / not perceived 50 was not clearly illuminated Light 100 was very bright 100 was dark 0 had normal conversation voices 50 had music Sound 50 had a vague background sound 100 was very noisy 100 was silent

114

115 Appendices

Appendix C. Statements and their values for the Event Profile Event spider components with their exact statements (single choice) and corresponding values:

0 = not relevant, 50 = important / positive emotions / past time; 100 = very important / negative emotions / future

For you, how important are the following categories in this event / dream scene?

Component Value Description

0 not relevant Actors (persons, animals, 25 somewhat important other living creatures) 50 important 100 very important 0 not relevant 25 somewhat important Objects 50 important 100 very important 80 negative 60 rather negative 0 not relevant Emotions valence +/- 100 both positive and negative emotions 20 rather positive 40 positive 25 weak 0 not relevant Emotions arousal 50 rather strong emotions 100 very strong emotions 100 failure 75 partial failure Goal (active participation) 0 no goal / not relevant 25 partial success 50 success 100 misfortune 75 partial misfortune Good Fortune or Misfortune 0 not relevant (passive) 25 partial good fortune 50 good fortune

116 50 in historical times 25 in my past 0 not defined Time 10 current time 75 in the near future 100 in the far future 0 not relevant 25 somewhat important Settings (location) 50 important 100 very important

117 Acknowledgments First of all, I would like to sincerely thank my doctoral supervisor Prof. Lorenz Hurni for sup- porting me with this challenging project idea and for giving me the opportunity to carry out this research. I am very grateful for his scientific advice and his valuable inputs during the whole project.

I would like to express my gratitude also to Prof. Jacques Montangero for the enlightening discussions regarding dream research, for his advices and helpful remarks.

Many thanks go to my co-examiner, Prof. William Cartwright, who showed vivid interest in this project and agreed to fly round half of the world to attend the exam.

I would like to sincerely thank Dr. Hansruedi Bär for his insightful comments, stimulating discussions and motivating words for the last stage of my thesis.

Many thanks go to Matthias Beilstein, for testing the developed data model, for sharing his mapping ideas and dreams with me.

I would like to also thank my colleagues at the Institute of Cartography and Geoinformation for their support and good talks about dreams and beyond. In particular, I am grateful to Raimund Schnürer and Dr. Magnus Heitzler for their advices and an open ear, to Marianna Farmakis and Wenke Zimmermann for sharing and discussing their dreams with me, to Claudia Matthys, Nadia Panchaud and Corina Pachlatko for motivating me.

I would like to thank to my family and my friends for their interest in my work, for their motivating words and appreciation. I am especially grateful to my husband and former colleague Dr. Ionuț Iosifescu Enescu for his support, encouragement, for believing in me and for always taking his time to listen to and discuss with me (also) the scientific aspects of this research.

This research was funded by the Swiss National Science Foundation, grant no. 205121_157087.

118