UNESP – UNIVERSIDADE ESTADUAL PAULISTA  

“JÚLIO DE MESQUITA FILHO” 

CÂMPUS DE PRESIDENTE PRUDENTE 

Programa de Pós-graduação em Ciências Cartográficas 

 

 

 

GUSTAVO GRASSI 

 

 

 

COLOR PREFERENCE OF CARTOGRAPHIC SYMBOL DESIGN RELATED TO 

THE URBAN LANDSCAPE CHARACTERISTICS OF SÃO PAULO WESTERN 

REGION FOR 1:10,000 TOPOGRAPHIC MAPS 

 

 

 

 

 

 

 

 

 

 

Presidente Prudente – SP 

2021



2 
 

  

UNESP – UNIVERSIDADE ESTADUAL PAULISTA  

“JÚLIO DE MESQUITA FILHO” 

CÂMPUS DE PRESIDENTE PRUDENTE 

Programa de Pós-graduação em Ciências Cartográficas 

 

 

 

 

 

 

GUSTAVO GRASSI 

 

 

COLOR PREFERENCE OF CARTOGRAPHIC SYMBOL DESIGN RELATED TO 

THE URBAN LANDSCAPE CHARACTERISTICS OF SÃO PAULO WESTERN 

REGION FOR 1:10,000 TOPOGRAPHIC MAPS 

 

 

 

 

 

 

 

 

 

 

Presidente Prudente – SP 

2021 

  

Master Science dissertation presented to the 

Programa de Pós-graduação em Ciências 

Cartográficas, Faculdade de Ciências e 

Tecnologia de Presidente Prudente/SP – UNESP. 

Supervisor: Prof. Dr. Edmur Azevedo Pugliesi  



3 
 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sistema de geração automática de fichas catalográficas da Unesp. Biblioteca da Faculdade 

de Ciências e Tecnologia, Presidente Prudente. Dados fornecidos pelo autor(a). 

 

Essa ficha não pode ser modificada. 

G769c 
Grassi, Gustavo 

Color preference of cartographic symbol design related to the urban 

landscape characteristics of São Paulo western region for 1:10,000 

topographic maps / Gustavo Grassi. -- Presidente Prudente, 2021 

145 p. 

Dissertação (mestrado) - Universidade Estadual Paulista (Unesp), 

Faculdade de Ciências e Tecnologia, Presidente Prudente  

Orientador: Edmur Azevedo Pugliesi 

1. Topographic map. 2. Cartographic design. 3. Cartographic 

symbol design. 4. Landscape. 5. Color preference. I. Título. 



4 
 

  

 

  



5 
 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

To my beloved parents, Zilda and Orlando, to my 

sister, Júlia, to my cats (Sami, Ninim, especially 

Ziva), to my dogs (Kate and Mulanguim), in 

memoriam of Lilica. 

To my sensei, dr. Daisaku Ikeda. 

To my great friends, Bruno, Mari, Vito, and 

colleagues with whom I shared this brief journey. 



6 
 

  

 ACKNOWLEDGEMENTS 

 

This study was funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico – 

CNPq. Grant: 132401/2019-3. 

 

I would like to thank the UNESP (Universidade Estadual Paulista) – Faculdade de Ciências e 

Tecnologia, Câmpus de Presidente Prudente and PPGCC (Programa de Pós-graduação em 

Ciências Cartográficas) for all support and infrastructure to develop this research.  

 

I am thankful to my supervisor, Prof. Dr. Edmur Azevedo Pugliesi, for his guidance and support 

in this research. 

 

I am thankful to my friends and colleges from PPGCC for their support and collaboration in 

developing this work. 

 

I am thankful to all those who kindly agreed to participate in the research as subjects. 

 

This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível 

Superior - Brasil (CAPES) - Finance Code 001.  

 

  



7 
 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

“Buddhism is a teaching of unsurpassed reason. 

Therefore, the strength of your faith must manifest 

itself in the form of studying, exercising your 

ingenuity, and making twice as much effort as 

anyone else. Earnest daimoku is the wellspring for 

the energy to challenge these things.” 

Daisaku Ikeda  



8 
 

  

Abstract 

  

This research aims to evaluate the preference for buildings cartographic symbol designs for 

topographic maps at 1:10,000 scale, to define symbols that may represent part of the western 

urban landscape located in the state of São Paulo. Three proposals were designed to represent 

urban buildings in the context of topographic maps: Pink visually coherent with the national 

cartographic conventions; Orange visually coherent with the urban landscape of the western 

portion of the São Paulo state; Gray visually coherent with the digital navigational maps used 

frequently. The map evaluation considered the preference for the symbol designs by using two 

groups of participants, one named architects, and the other named non-architects. Participants 

gave their opinions about the designs, scored, and ordered them as well. The Orange design, 

understood as visually coherent with the urban landscape of the western portion of the São 

Paulo state, was the most preferred in all aspects, in terms of opinion, score, and order of 

preference. Despite that, the differences in score and order of preference between Orange and 

Pink designs were not statistically significant. On the other hand, the differences in score and 

order of preference between Gray and the two other designs were statistically significant, 

indicating that the Gray as the least preferred. The participants' opinions indicated that the 

preference for the Orange is related to the better contrast between the buildings and the other 

cartographic symbols on the map. Also, the preference was influenced by the visual comfort 

provided by the composition, for being more colorful and more attractive, when compared with 

Pink and Gray. Furthermore, from the statistical analysis considered, map users with 

background knowledge in architecture and those with other backgrounds could be considered 

part of the same population. 

Keywords: Topographic map; Cartographic design; Cartographic symbol design; Landscape; 

Color preference. 

  



9 
 

  

Resumo 

 

O objetivo desta pesquisa é avaliar a preferência de símbolos cartográficos de edificações em 

mapas topográficos na escala 1: 10.000, para definir símbolos que podem representar parte da 

paisagem urbana localizada no oeste do estado de São Paulo. Foram elaboradas três propostas 

de mapas topográficos para representação de edifícios urbanos: Rosa visualmente coerente com 

as convenções cartográficas nacionais; Laranja visualmente coerente com a paisagem urbana 

do oeste paulista; Cinza visualmente coerente com os mapas de navegação digitais usados com 

frequência. A avaliação do mapa considerou a preferência pelos desenhos dos símbolos usando 

dois grupos de participantes nomeados de arquitetos e não arquitetos. Os participantes deram 

suas opiniões sobre os designs, pontuaram e também os ordenaram. O design Laranja entendido 

como visualmente coerente com a paisagem urbana do oeste paulista foi o mais preferido em 

todos os aspectos, ou seja, em termos de comentários, pontuação e ordem de preferência. Apesar 

disso, as diferenças em pontuação e ordem de preferência entre os designs Laranja e Rosa não 

foram estatisticamente significativas. Por outro lado, as diferenças na pontuação e ordem de 

preferência entre Cinza e os dois outros designs foram estatisticamente significativas, indicando 

que o Cinza foi o menos preferido. As opiniões dos participantes indicaram que a preferência 

pelo Laranja está relacionada com o melhor contraste entre as edificações e os demais símbolos 

cartográficos no mapa. Além disso, a preferência foi influenciada pelo conforto visual 

proporcionado, por ser mais colorido e atraente que o Rosa e o Cinza. Além disso, a partir da 

análise estatística utilizada nesta pesquisa, os usuários do mapa com formação em arquitetura 

e aqueles com outras origens podem ser considerados parte da mesma população.   

Palavras-chave: Mapa topográfico; Projeto cartográfico; Projeto de símbolo cartográfico; 

Paisagem; Preferência por cor. 

  



10 
 

  

Figure Index 
 

 

Figure 1 – Electromagnetic spectrum ....................................................................................... 25 

Figure 2 – Human eye components .......................................................................................... 26 

Figure 3 – Grouping by color similarity, white circles versus black circles ............................ 28 

Figure 4 – Difference between the perception of the figure and ground as a function of the 

contour: a) weak contour refers to the weak perception of the ground; b) better differentiation 

between the figure and the ground by highlighting the outline ................................................ 29 

Figure 5 – Example of simultaneous contrast or induction ...................................................... 32 

Figure 6 – Example of successive contrast or color differentiation ......................................... 33 

Figure 7 – Workflow for creating the three designs ................................................................. 36 

Figure 8 – Area of study in the São Paulo County highlighted by black contour selected for 

this research .............................................................................................................................. 38 

Figure 9 – Clippings of satellite images in the true color composition of cities in the São 

Paulo state ................................................................................................................................ 44 

Figure 10 – Clippings of satellite images in true color composition from the capital and 

metropolitan region of the São Paulo state ............................................................................... 45 

Figure 11 – Clippings from municipalities of the western of the São Paulo state: a, b, c, d, e, f 

satellite images in true color composition, 2020 g, h, i, j, k, l orthoimages at 1:10,000 scale, 

2010 .......................................................................................................................................... 46 

Figure 12 – Clippings of digital navigational maps: Google Maps above and Waze online 

bellow ....................................................................................................................................... 47 

Figure 13 – Clippings of proposed designs applied for the area 01 at 160 dots per inch 

resolution, zoom 100%: Pink (on the top), Orange (in the middle), Gray (at the bottom) ...... 50 

Figure 14 – Proposed designs applied for the area 01: Pink (on the top), Orange (in the 

middle), Gray (at the bottom) ................................................................................................... 51 

Figure 15 – Proposed designs applied for the area 02: Pink (on the top), Orange (in the 

middle), Gray (at the bottom) ................................................................................................... 52 

Figure 16 – Proposed designs applied for the area 03: Pink (on the top), Orange (in the 

middle), Gray (at the bottom) ................................................................................................... 53 

Figure 17 – Proposed designs applied for the area 04: Pink (on the top), Orange (in the 

middle), Gray (at the bottom) ................................................................................................... 54 

Figure 18 – Proposed designs applied for the area 05: Pink (on the top), Orange (in the 

middle), Gray (at the bottom) ................................................................................................... 55 

Figure 19 – Proposed designs applied for the area 06: Pink (on the top), Orange (in the 

middle), Gray (at the bottom) ................................................................................................... 56 

Figure 20 – Proposed designs applied for the area 07: Pink (on the top), Orange (in the 

middle), Gray (at the bottom) ................................................................................................... 57 

Figure 21 – Workflow of the steps used to evaluate the user’s preference for the proposed 

designs ...................................................................................................................................... 58 

Figure 22 – Steps used during the interview with the questionnaires ...................................... 63 

Figure 23 – Boxplot showing means for each design, according to the preference of architects 

in (a) and non-architects in (b) ................................................................................................. 76 

Figure 24 – Boxplot showing the means by design assessed by architects and non-architects 77 

Figure 25 – Mean of preference responses from the group of (a) architects (b) and non-

architects ................................................................................................................................... 79  



11 
 

  

Table Index 
 

 

Table 1 – Specification of the feature classes selected for this work ....................................... 39 

Table 2 – Graphical specification of some feature classes presented in the T34-700 manual . 42 

Table 3 – Graphical specification of the cartographic symbols ............................................... 48 

Table 4 – Architects' opinions about the Pink design .............................................................. 68 

Table 5 – Architects' opinions by feature class in the Pink design .......................................... 68 

Table 6 – Architects' opinions about the Orange design .......................................................... 69 

Table 7 – Architects' opinions by feature class in the Orange design ...................................... 69 

Table 8 – Architects' opinions about the Gray design .............................................................. 70 

Table 9 – Architects' opinions by feature class in the Gray design .......................................... 70 

Table 10 – Non-architects’ opinions about the Pink design ..................................................... 71 

Table 11 – Non-architects' opinions by feature class in the Pink design ................................. 71 

Table 12 – Non-architects’ opinions about the Orange design ................................................ 72 

Table 13 – Non-architects' opinions by feature class in the Orange design ............................. 72 

Table 14 – Non-architects’ opinions about the Gray design .................................................... 73 

Table 15 – Non-architects' opinions by feature class in the Gray design ................................ 73 

Table 16 – Summary of the statistical distribution employed to the mean of the datasets for 

the rating scores assigned by architects and non-architects. The significance level considered 

was 5% (≥ 0.05) ....................................................................................................................... 75 

 

  



12 
 

  

Summary 

 

1. Introduction ................................................................................................................... 14 

1.1. Aim ........................................................................................................................ 18 

1.2. Justification ............................................................................................................ 19 

1.3. Dissertation overview ............................................................................................ 20 

2. Literature Review .......................................................................................................... 21 

2.1. Topographic Maps ................................................................................................. 21 

2.2. Landscape .............................................................................................................. 23 

2.3. Light and receptors ................................................................................................ 24 

2.4. Perceptual organization .......................................................................................... 26 

2.5. Principles of colors for cartography ....................................................................... 29 

2.5.1. Color vision ........................................................................................................ 29 

2.5.2. Color components and color models .................................................................. 30 

2.5.3. Color contrast and differentiation ...................................................................... 31 

2.5.4. Subjective reactions to color .............................................................................. 33 

2.6. Map evaluation ....................................................................................................... 34 

3. Cartographic design for the topographic maps ............................................................. 36 

3.1. Planning the map .................................................................................................... 37 

3.2. Preparing the cartographic database ...................................................................... 38 

3.3. Designing cartographic symbols for the proposed topographic maps ................... 39 

3.3.1. Visual coherence between the proposition and the national cartographic 

convention ......................................................................................................................... 41 

3.3.2. Visual coherence between the proposition and the western landscape in the state 

of São Paulo ...................................................................................................................... 43 

3.3.3. Visual coherence between the proposition and the digital navigational maps 

used frequently .................................................................................................................. 46 

3.4. Proposed designs and graphical specifications ...................................................... 48 

4. Assessment of the user’s preference ............................................................................. 58 

4.1. Preparing the questionnaire and commitment documents ..................................... 59 

4.2. Subjects .................................................................................................................. 59 

4.3. Experimental design ............................................................................................... 60 

4.4. Procedure ............................................................................................................... 61 

5. Results ........................................................................................................................... 64 

5.1. Analysis of the opinions regarding the characteristics of the designs ................... 64 

5.1.1. General analysis by the design composition ...................................................... 65 



13 
 

  

5.1.2. Analysis by toponymy and individual cartographic symbols ............................ 66 

5.2. Analysis of the rating scores assigned to the designs ............................................ 74 

5.2.1. Statistical analysis of the data distribution ......................................................... 74 

5.2.2. Comparison of the scores for architects and non-architects ............................... 75 

5.2.3. Comparison of the scores between architects and non-architects ...................... 76 

5.3. Analysis of the order of preference for the designs ............................................... 77 

5.3.1. Percentage of the order of preference ................................................................ 78 

5.3.2. Paired comparison for architects and non-architects .......................................... 79 

5.3.3. Paired comparison between the architects and non-architects ........................... 80 

5.4. Discussion .............................................................................................................. 81 

6. Conclusions and recommendations ............................................................................... 86 

REFERENCES ..................................................................................................................... 90 

APPENDIX A – DOCUMENTS FOR THE ETHICAL AND LEGAL PURPOSE 

PROVIDED FOR THE LOCAL RESEARCH AND ETHIC COMMITTEE ..................... 96 

APPENDIX B – QUALITATIVE DATA: SUBJECTS’ RESPONSES FOR WHAT THEY 

THINK ABOUT THE DESIGNS ...................................................................................... 106 

APPENDIX C – QUANTITATIVE AND ORDINAL DATA: SUBJECTS’ RESPONSES 

ABOUT SCORE RATING AND PREFERENCE ORDERING THE DESIGNS ............ 133 

APPENDIX D – CATEGORIZATION OF DESCRIPTIONS REPORTED BY SUBJECTS 

IN TERMS OF AESTHETIC DIMENSIONS AND USABILITY ITEMS ...................... 142 

 

  



14 
 

  

1. Introduction  

 

Topographic maps represent natural and anthropic geographic features of the 

Earth’s surface (KEATES, 1973). These kinds of maps present a wide variety of landscape 

features, and they serve as a basis for several applications (ROBINSON et al., 1995; SSC, 2005; 

KENT, 2009; COLLIER, 2009). The beauty associated with topographic maps is related to how 

the landscape characteristics are symbolized cartographically (KENT, 2005). Each country has 

its symbology style, which is applied to the national topographic maps (ROBINSON et al., 

1995) and the style may characterize a national identity (KENT, 20009; KENT; VUJAKOVIC, 

2009). 

Studies regarding the representation of topographic maps that have been conducted 

worldwide are limited (SLUTER et al., 2018). Some of the research in Europe focused on 

understanding the characteristics and peculiarities of the topographic map's styles used by the 

national mapping agencies in European countries (KENT, 2009; KENT; VUJAKOVIC, 2009, 

2011; ORY et al., 2015). The style of a topographic map could be identified by symbology and 

texts, which is employed to describe landscape (KENT, 2005). Those works considered 

topographic maps at 1:50,000 scale to identify stylistic similarities between national 

symbology. 

In the United States, Raposo and Brewer (2014) studied the preference of different 

cartographic designs by using orthoimages as background for U.S. topographic maps at 

1:24,000 scale. They tested different designs for the whole U.S. considering several different 

landscapes. Their results identified that the preference for the topographic map design was 

influenced by the landscape portrayed. 

In Brazil, efforts have been engaged on studying large-scale topographic maps 

mainly with cartographic generalization (COMÉ; SLUTER, 2015; CASTRO; SLUTER, 2019; 

GRASSI; PUGLIESI, 2020b), cartographic representation at multiple scales (MENDONÇA; 



15 
 

  

AMORIM; PUGLIESI, 2017), and cartographic symbol design (NATINGUE, 2014; SLUTER 

et al., 2018; LEMES NETO, 2020). Brazilian researchers have, recently, concentrated on 

studying large-scale topographic maps representations due to the absence of cartographic 

conventions1 (SLUTER et al., 2018). 

Paraná state established its large-scale cartographic conventions visually coherent 

with the national cartographic conventions (SLUTER et al., 2018). In São Paulo, cartographic 

conventions are established by Instituto Geográfico e Cartográfico do Estado de São Paulo - 

IGCSP (Geographic and Cartographic Institute of the São Paulo State), a state mapping agency. 

Among the products made and supplied by IGCSP, one of them is the topographic maps at 

1:10,000 scale, which are resulting from photogrammetric restitution by using Computer-Aided 

Design (CAD) tools. Observing the urban landscape of the western region of the São Paulo 

state through a true color composition of satellite images or orthoimages at 1:10.000 scale, some 

features are noticeable: buildings in a mix of orange, red, and brown shades; paved roads in 

dark gray; and vegetation in medium-dark green shades. Those features seem to be 

representative of the urban space at 1:10.000 scale. 

Both state agencies mentioned previously employ different ways of representing 

the urban landscape. It seems that both of them seek to follow visual coherence with the 

conventions established by the national topographic mapping2 agency instead of coherence with 

certain features of the urban landscape, such as those buildings symbols in shades of red. 

According to Sluter et al. (2018), in Parana state the Câmara Técnica de Cartografia e 

Geoprocessamento – CTCG (Technical Chamber of Cartography and Geoprocessing) agency 

requested from their research group a study about large-scale topographic map conventions. 

They declared have followed principles of legibility for topographic maps which were 

 
1 The definition of cartographic conventions on topographic maps follows Kent and Vujakovic (2009),  “e.g. 
surronding the use of color, such as blue for water, brown for contours, green for vegetation, and black for 
‘cultural’ features”.  
2 In Brazil, the expressions “national systematic mapping” or “systematic mapping” are used for topographic 
mapping. 



16 
 

  

presented by Keates (1973). The result presented by the authors is similar to the national 

cartographic conventions. 

On the other hand, to represent the portion west of the São Paulo state on large-

scale maps, designing cartographic symbols and evaluating the communication quality of those 

symbols are necessary to recommend to the IGCSP agency appropriate ways of characterizing 

the landscape. Since this study is purely scientific, were sought options and non-orthodoxy 

possibilities for conceiving a large-scale topographic map representation, which is the first 

Brazilian study to explore this unknown area.  

The landscape representation through topographic map symbols could be 

established based on different aspects of visual coherence, such as the conventions established 

for the national topographic mapping, the experience or habit of using online maps for 

navigation, the similarity of the symbol with the regional landscape. The national cartographic 

conventions that represent the official topographic maps, for scales between 1:25,000 and 

1:250,000, are available in the technical manual T34-700, are established by the Diretoria de 

Serviço Geográfico do Exército (Brazilian Army) (BRASIL, 2000). In the absence of 

cartographic conventions for large-scale topographic maps, the municipalities or the states 

could have available a standard symbology to produce a homogeneous mapping for all 

municipalities that respect the regional and local landscape characteristics. 

The frequent experience or the habit of using online navigational maps (such as 

Google Maps, Bing Maps, ArcGIS Online, or Waze) may influence the preferences when 

choosing the symbolized landscape through a topographic map. Experience or habit is a Gestalt 

law in searching for visual patterns (ELLIS, 1955; MACEACHREN, 1995). According to 

Dearden (1984), familiarity and habit with particular places, such as forests, farms, or cities in 

recreational activities and travels, influenced the landscape preference. Additionally, Hammitt, 

Knauf, and Noe (1989) found that recreational activity preference is strongly related to 

experience in that activity, not their expertise. 



17 
 

  

The beauty and aesthetic of a place can be translated into cartographic conventions 

and applied to topographic maps (KENT 2005). Associating the cartographic symbol with the 

landscape could define identities that vary through the space on a topographic map. Thus, the 

cartographic convention could be thought of and developed to portray a local or a regional 

landscape. 

The landscape perception is strongly related to the scale and to the observer’s 

cultural background, and "it could be defined as the domain of the visible, that which the view 

encompasses" (SANTOS, 1988). Considering individual differences, people with different 

characteristics or levels of knowledge, such as a professional trained in architecture, an artist, 

or an ordinary person, could perceive the landscape differently (SANTOS, 1988). 

The process of perceiving the local landscape that is portrayed on a topographic 

map may be entirely dissimilar among readers with different levels of knowledge. In addition, 

as the topographic map symbols would be consistent with the local landscape, the readers’ 

preference for the landscape may influence his preference for different map designs. 

In research carried out by Dearden (1984), the landscape preference of a group of 

participants was strongly influenced by the current living environment (last five years) and past 

landscape experiences. Also, the recreational activities and travel, otherwise, professional 

expertise and socio-economic variables did not influence them. In the matter of topographic 

map design preference, Raposo and Brewer (2014) found out that the landscape characteristics 

had direct influences on the map readers' preference. 

Consequently, the landscape of the living environment seems to affect the 

individuals' preference for the landscape. It appears that the preference for different topographic 

map designs is closely related to the landscape characteristics. Although the national 

conventions are related to the landscape, the generic character of the landscape for the whole 

country may not be the most appropriate at the local and regional levels. 



18 
 

  

Hence, two questions are presented by considering the visual coherence between 

the propositions created in this research and the concepts that they represent, intending to supply 

the absence of large-scale topographic map conventions for the São Paulo state. First, which 

kind of design best represents the landscape of the western region of the state of São Paulo: 

national conventions, online maps for navigation, or urban landscape? Second, could 

professional training influence the preference for choosing different designs?  

It is hypothesized that people who live in the west of the São Paulo state will prefer 

topographic maps at 1:10,000 scale more visually related to the landscape characteristics that 

are present in the region where they live due to previous experience with that landscape rather 

than a design visually coherent with the national cartographic conventions, or visually coherent 

with the online navigation maps. 

 

1.1. Aim 

 

This research aims to evaluate the preference for cartographic symbol designs that 

represent buildings on topographic maps at 1:10,000 scale, to define symbols that may represent 

part of the western landscape of the state of São Paulo. The specific objectives for this study 

are summarized in: 

- Design three proposals of buildings cartographic symbol for topographic maps 

considering three types of visual coherence: national conventions, online maps 

for navigation, and urban landscape. 

- Analyze the subject’s opinions about each proposed design. 

- Analyze the subject’s preference for the designs in terms of score and order of 

preference. 

 



19 
 

  

1.2. Justification 

 

The topographic maps at 1:10,000 scale, produced by the Instituto Geográfico e 

Cartográfico do Estado de São Paulo – IGCSP (Geographic and Cartographic Institute of the 

São Paulo State), are available in CAD (Computer-Aided Design) and paper formats. However, 

those types of material have their representations limited for database purposes and for viewing 

in digital media. In contrast, some countries offer online platforms for topographic maps, such 

as the United States of America (United States Geological Survey – USGS), Germany 

(Geodateninfrastruktur Deutschland – GDI), the United Kingdom (Ordnance Survey – OS), 

and the Netherlands (Publieke Dienstverlening Op de Kaar – PDOK), also due to the digital 

technology advances and a stable economy. 

In Brazil, an initial effort has been conducted 

(https://bdgex.eb.mil.br/bdgexapp/mobile/), but it seems that official national topographic maps 

are far away to be constructed and used effectively online in the following years. Since there 

are no national cartographic conventions for maps at 1:10,000 scale, topographic maps 

produced by IGCSP should have their conventions. 

Kent (2009) points out the importance of representing topographic maps as a means 

of revealing a “socially constructed landscape,” which could be understood as “achievement of 

political independence”. In addition to portraying the landscape, the cartographic symbols could 

reveal its beauty (KENT, 2005). A beautiful, pleasant, and attractive map tends to be used more 

often. In the digital environment, especially on the internet, the beauty and quality of the map 

are vital elements to attract the user to read, return and continue visiting your content (DEEB 

et al., 2015).   

Topographic maps can be used to implement several applications that are part of 

people's daily lives. It is possible to think that the availability of topographic maps on the 

internet that reveal the landscape features can be reached by many users. In this sense, the 



20 
 

  

construction of a symbology based on theories of cartographic communication and psychology 

of perception can contribute to achieving map communication and encourage the use of 

topographic maps. 

When considering the state of São Paulo, the most populous in Brazil, the 

development of user-oriented symbology can positively impact many users in different areas, 

such as education, urbanism, business, and engineering. This work aims to take a step towards 

those paths and seek an initial comprehension about the topographic maps preference with a 

group of people who live in the west of the São Paulo state and serve as a source of inspiration 

and reference for further studies research on this subject. 

 

1.3. Dissertation overview 

 

This dissertation was structured in the following five chapters. Chapter 2 

“Literature review” serves as the theoretical basis for elaborating and evaluating the 

cartographic designs: topographic maps, landscape, light and receptors, perceptual 

organization, principles of color for mapping, and map evaluation. Chapter 3 “Cartographic 

design for the topographic maps” considers the processes of planning the map, obtaining and 

preparing the cartographic database, designing the cartographic symbols. Chapter 4 

“Assessment of the user’s preference” presents the steps employed to prepare questionnaires, 

and commitment documents that are required for the interviews, the recruitment of subjects, the 

experimental design, and the details of the procedure that involves the map evaluation. Chapter 

5 “Results” describes the process of analyzing qualitative and ordinal data obtained from the 

interviews with the subjects, taking into account the statistical distribution of the data, and the 

type of statistical analysis that fit with the data. Chapter 6 “Conclusion and recommendations” 

presents the analysis of the results and bring out ideas for further research. 

  



21 
 

  

2. Literature Review 

2.1. Topographic Maps 

 

Topographic maps are directly related to the topography, which refers to 

representing and identifying all features that are located on the Earth's surface, whether a natural 

or anthropic character, always associated with a specific position (KEATES, 1973).  What will 

be portrayed on the map occurs by the importance of the features in its geographical context 

(COLLIER, 2009). The SSC (2005) indicates that no one feature should stand out strongly to 

others, except in specific situations.  

The employment of cartographic conventions is essential for understanding 

topographic maps (ROBINSON et al., 1995). The cartographic conventions on the map may 

reveal the beauty of the landscape that is being represented (KENT, 2005). Each country has 

its map agency that is responsible for organizing the cartographic conventions (ROBINSON et 

al., 2009; COLLIER, 2009). Although the symbology employed for topographic maps varies 

from country to country, the geographic features represented do not differ much (COLLIER, 

2009; KENT, 2009; KENT; VUJAKOVIC, 2011).  

The focus of research on topographic maps representations in the world and Brazil 

are identified. Worldwide, the focus has been on understanding the different styles and 

particularities that are presented on the map face (KENT, 2009; KENT; VUJAKOVIC, 2009; 

KENT; VUJAKOVIC, 2011; ORY et al., 2015), as well as on users' preference over the maps 

(RAPOSO; BREWER, 2014).  

Kent (2009) offered an initial approach to establish a methodology for analyzing 

topographic map styles based on exploring a thematic map and topographic map symbols. In 

his work, it is considered some European countries, and he focused on the changes brought by 

the technological and cultural advances over time. Kent and Vujakovic (2009) evaluated the 



22 
 

  

graphical characteristics presented on topographic maps of some European countries, and they 

found a unique topographic map style that characterizes national identities. 

Two years later, Kent and Vujakovic (2011) identified the adoption of a 

supranational and unique symbology as a challenge since the meaning and cultural values 

attributed to the symbols on topographic maps remain relevant. Instead of supporting just a 

unique standard to the European countries, they suggested two levels of representation, one 

national and the other supranational, and the users could choose that. 

Seeking to understand whether French and Swiss people could differentiate or not 

between topographic map styles of those countries, Ory et al. (2015) found that most people 

could distinguish the styles based on the toponymy, relief, and geographical landscape 

characteristics. Considering other kinds of representing topographic maps, Raposo and Brewer 

(2014) inquired people about their preference and map readability when using U.S. topographic 

map designs with orthoimages. They identified which landscape had a significant influence 

over the preference. However, readability did not seem influenced by the landscape. 

In Brazil, the large-scale topographic map remains without a cartographic 

convention at the national, state, or local levels. Once DSG establishes scales between 1:25,000 

to 1:250,000, and minors than 1:250,000 are established by Instituto Brasileiro de Geografia e 

Estatística – IBGE (Brazilian Institute of Geography and Statistics). Large-scale topographic 

maps conventions must be established by municipalities or state mapping agencies (SLUTER 

et al., 2018). Some Brazilian researchers have focused on cartographic generalization (COMÉ; 

SLUTER, 2015; MENDONÇA; AMORIM; PUGLIESI, 2017; CASTRO; SLUTER, 2019; 

GRASSI; PUGLIESI, 2020b) and cartographic symbol design (NATINGUE, 2014; LEMES 

NETO, 2020; SLUTER et al., 2018).  

Natingue (2014) proposed symbols to represent geographic features on topographic 

maps at 1:5,000 scale for the Parana state. Comé and Sluter (2015) suggested a technique to 

detect geometrical conditions that could be appropriate for cartographic generalization for 



23 
 

  

buildings, urban roads, and property boundaries, considering scales between 1:2,000 and 

1:10,000. Mendonça, Amorim, and Pugliesi (2017) developed a graphic representation of a 

zoomable and interactive topographic map, working with scales between 1:10,000 and 

1:25,000. 

Sluter et al. (2018) carried out a bibliographic survey about topographic maps in 

Brazil and identified a lack of research work in the national context. Also, they defined 

geographic features that should be part of the 1:2,000 scale, as well as the symbology for those 

features. The results also are recommendations for the Câmara Técnica de Cartografia e 

Geoprocessamento – CTCG (Technical Chamber of Cartography and Geoprocessing), located 

in Parana state. 

Castro and Sluter (2019) aimed to establish minimum dimensions for a group of 

cartographic symbols for topographic maps that represent urban features by using colored area 

and linear symbols. Lemes Neto (2020) evaluated the readability of the symbology established 

in the technical manual T34-700, as well as a group of proposed symbols, both for coffee and 

sugar cane. Grassi and Pugliesi (2020b) studied cartographic generalization for blocks, 

buildings, urban and rural road features considering the scales between 1:10,000 and 1:50,000. 

Studies that explore other possibilities for large-scale topographic map representations beyond 

in line with the national cartographic convention were not found. 

 

2.2. Landscape 

 

The landscape may be defined as everything that belongs to the domain of the 

visible (SANTOS, 1988). The visual connotation of the landscape is not restricted to a single-

framed view or an aesthetic pleasure, and it is also related to a more significant concern with 

viscerally and experience (MORIN, 2009). The concept of landscape linked to people's 

perception is related to the tradition of characterizing the landscape with roots in art and 



24 
 

  

humanity (SIMENSEN; HALVORSEN; ERIKSTAD, 2018). The landscape perception is 

strongly related to the observation scale and the observer's cultural background (SANTOS, 

1988). 

People with different knowledge perceive the landscape differently (SANTOS, 

1988). However, although a background knowledge in architecture may influence landscape 

perception, Dearden (1984) points out that it does not strongly relate to landscape preference. 

The landscape preference is more related to aspects of familiarity or habit, such as previous 

experience with the landscape, where the person is living for the last 5 years, their recreational 

activities and travel, and activities of personal engagement (DEARDEN, 1984). 

Regarding the urban landscape perception, the studies conducted by Lynch (1960) 

and Downs and Stea (1973) may help to understand urban space from people's mental images. 

Lynch (1960) identified five essential elements that people use to orient themselves in North 

American cities: paths, edges, districts, nodes, and landmarks. Downs and Stea (1973) relate 

cognitive images to behavior patterns, and they identified that mental images are constructed 

as a function of activities performed by people.  

 

2.3. Light and receptors 

 

The electromagnetic spectrum is the organization and classification of the 

electromagnetic energy generated by the light as a function of the wavelength, commonly with 

the unit of measurement in nanometers or micrometers (DENT; TORGUSON; HOLDER, 

2009). The electromagnetic spectrum is classified into gamma rays, X-rays, ultraviolet, visible, 

infrared, microwave, radio waves, and long waves (HALLIDAY; RESNICK; WALKER, 

2009), as depicted in Figure 1. The portion of the spectrum perceived by the human visual 

system corresponds to the visible spectrum or light, allowing humans to sense colors. It 

corresponds to wavelengths of about 400 to 700 nm (ROBINSON et al., 1995). 



25 
 

  

 

Figure 1 – Electromagnetic spectrum 

 
From: Dent, Torguson, and Holder (2009) 

 

The human eye has the function of making the acquisition of light to be transferred 

and processed by the brain (ROBINSON et al., 1995). The pair of eyes allows the stereoscopic 

view of what is observed. Each eye has a spherical shape with approximately 2.5 cm in diameter 

(Figure 2). 

 

 

 

 

 

 

 



26 
 

  

 

 

Figure 2 – Human eye components 

 

From: Dent, Torguson and Holder (2009) 

 

In the retina (Figure 2), the fovea is where the more excellent visual acuity is found, 

the region of the optic nerve where communication takes place between the retina, and the 

brain; a region of the optic nerve that creates what is called a blind spot (SLOCUM et al., 2009). 

Cones and rods are found in the retina, which are two types of cells capable of recording the 

intensity of light, and the perception of wavelength will give the sensation of colors 

(ROBINSON et al., 1995; MACEACHREN, 1995). 

 

2.4. Perceptual organization 

 

Gestalt theory, dated to the early 20th century, is the result of the work conducted 

by German psychologists, including Wertheimer, Kohler, and Koffka (MACEACHREN, 

1995). Their work sought to establish an understanding of the visual field (PENNA, 2000). As 

stated by MacEachren, "the Gestalt approach emphasized the holistic nature of human reactions 

to sensation." Wertheimer affirms that “there are wholes, the behavior of which is not 



27 
 

  

determined by that of their elements, but where the part-processes are themselves determined 

by the intrinsic nature of the whole.” (ELLIS, 1955). 

Two essential aspects compound the Gestalt principles, linked to visual scene 

recognition, perceptual grouping, and figure-ground segregation (MACEACHREN, 1995), also 

known as two components of perceptual organization: grouping and segregation 

(GOLDSTEIN; BROCKMOLE, 2009). In the perceptual grouping, a particular group of 

elements presented in a scene is perceived as a group with similar characteristics or attributes. 

In other words, the elements are perceived as belonging to a single-family or different families 

according to their appearance (MACEACHREN, 1995). 

Wertheimer presents nine laws, or rules, that explain the occurrence of perceptual 

grouping (ELLIS, 1955): proximity, common fate, prägnanzstufen (concise forms), objective 

set, experience or habit, good continuation, simplicity, similarity, and closure. Proximity means 

that objects arranged nearby tend to be perceived as a unique structure (MACEACHREN, 1995; 

GOLDSTEIN; BROCKMOLE, 2009).  

The common fate means that objects moving together or in the same direction are 

seen as a group (MACEACHREN, 1995; GOLDSTEIN; BROCKMOLE, 2009). The concise 

forms represent a threshold of stability in which objects form grouping (ELLIS, 1955; 

MACEACHREN, 1995). The objective set indicates that if any change occurs, there will be a 

tendency to form stable groups (ELLIS, 1955; MACEACHREN, 1995). The experience or habit 

indicates that grouping occurs when observing familiar shapes or arrangements 

(MACEACHREN, 1995). 

The most critical perceptual grouping rules to be considered in this present study 

are good continuation, simplicity, similarity, and closure. A good continuation is achieved when 

elements that follow a constant direction group, such as points, traces, lines, or areas, remain 

continuous (MACEACHREN, 1995; GOLDSTEIN; BROCKMOLE, 2009). The simplicity or 

good figure means that objects will be grouped in the simplest form or figure (MACEACHREN, 



28 
 

  

1995; GOLDSTEIN; BROCKMOLE, 2009). As a visual technique, simplicity is presented by 

low quantities of visual units and it is characterized by visual organizations that are easily 

assimilated, read, and understood quickly (GOMES FILHO, 2015). The similarity indicates that 

elements of the same appearance are identified as groups (Figure 3) (MACEACHREN, 1995; 

GOLDSTEIN; BROCKMOLE, 2009). The closure shows that closed objects tend to be 

perceived as whole shapes and bounded objects. (MACEACHREN, 1995). 

The figure-ground segregation requires the visual sensor to organize the visual 

elements in two planes, one less important (ground or background) and the other which can 

attract the visual attention (MACEACHREN, 1995). The figure should attract the viewer's 

visual attention (GOLDSTEIN; BROCKMOLE, 2009), and according to MacEachren (1995), 

six rules work to achieve figure-ground segregation: heterogeneity, contour, surroundedness, 

orientation, relative size, and convexity.  

 

Figure 3 – Grouping by color similarity, white circles versus black circles 

 

From: Adapted from MacEachren (1995) 

 

The surroundedness means that objects surrounded tend to be seen as figures 

(BRUCE; GREEN, 1990; MACEACHREN, 1995). The orientation favor objects with 

horizontal or vertical orientation to be seen as figures, easier than diagonally oriented (BRUCE; 

GREEN, 1990; MACEACHREN, 1995). The relative size, named as size by Robinson et al. 

(1995), or positive and negative by Dondis (2003), indicates the smallest between two areas 

tends to be seen as a figure. Convexity means that elements whose contours have great 



29 
 

  

convexity tend to be seen as a figure more easily than elements with smooth edges 

(MACEACHREN, 1995). 

The main figure-ground segregation rules to be used in this research are 

heterogeneity and contour. The heterogeneity indicates that different elements must be present 

in the observed region (MACEACHREN, 1995).  Robinson et al. (1995) called this rule a 

differentiation factor. The contour considers that objects are more easily perceived when edges 

define them (MACEACHREN, 1995). This factor is also known as good contour (ROBINSON 

et al., 1995) or illusory contour (GOLDSTEIN; BROCKMOLE, 2009). The contour is directly 

related to the occurrence of heterogeneity or differentiation because even objects of similar 

coloring can be differentiated according to a contour (Figure 4). 

 

Figure 4 – Difference between the perception of the figure and ground as a function of the contour: a) 

weak contour refers to the weak perception of the ground; b) better differentiation between the figure 

and the ground by highlighting the outline 

 

From: Adapted from MacEachren (1995) 

 

2.5. Principles of colors for cartography 

2.5.1. Color vision 

 

The appropriate use of colors can positively improve visual search tasks, as pointed 

out in the study conducted by Williams (1967). In that work, it was identified that colors are 

more decisive in determining elements than shapes or size. In the context of web maps, 



30 
 

  

Muehlenhaus (2013) suggest the reduced use of colors and point out that the base map can be 

preferably white, light gray, dark gray, or black, thus avoiding other predominant colors. 

Color perception occurs by biochemical processes in the human brain, using 

electromagnetic energy captured by the eyes and transmitted by electrical pulses to the brain 

(WADE; SWANSTON, 2001; DENT; TORGUSON; HOLDER, 2009; SLOCUM et al., 2009). 

The eyes have two types of cells, cones, and roads. Cones are responsible for the decomposition 

of colors and the rods to light acquisition (WADE; SWANSTON, 2001). There are three types 

of cones, and each one responds to capture an electromagnetic wave interval, short (blue), 

medium (green), and long (red) (SLOCUM et al., 2009). 

Two main theories describe how electromagnetic energy is received by the eye and 

conveyed to the brain: trichromatic theory and theory of the opponent process. Trichromatic 

colors theory was developed by Thomas Young in 1801 and defended by Hermann von 

Helmholtz in 1852. This theory indicates that the transmission of electromagnetic waves 

happens directly from cones to the brain. 

The opponent-process theory was proposed by the physiologist Ewald Hering, in 

1877. It is based on the premise that the electromagnetic pulse is not transmitted directly to the 

brain, although there are cone cells. The cones capture the energy that is sent to ganglion cells 

(WADE; SWANSTON, 2001). The ganglia are characterized into three channels, one of gray 

intensity and two opposing color channels, one red-green and one blue-yellow (WADE; 

SWANSTON, 2001). Then, it works by mixing the pairs so that the brain perceives colors. The 

comprehension of color vision contributes to the knowledge about color components and 

additive color models, which, in turn, make a constant presence in building a colorful graphic 

design. 

 

2.5.2. Color components and color models 

 



31 
 

  

Colors can be understood from three components, or dimensions, that form the basis 

for additive and subtractive theories. The three components are hue, value or brightness, and 

saturation or chroma (ROBINSON et al., 1995; SLOCUM et at., 2009; MUEHLENHAUS, 

2013). The nomenclature of the components varies according to the model adopted (SLOCUM 

et al., 2009). 

Among the various color additive models, two are very noticeable due to their 

presence in a GIS environment, the RGB and the HSV models. The RGB color additive model 

is machine-oriented, as it performs the specifications in red, green, and blue light values 

(SLOCUM et al., 2009). This model can be represented by a color cube, which along a central 

diagonal represent the values of gray that vary from black to white, and in each main axis that 

composes the prominent faces of the cube, each of the components materializes red, green, and 

blue (DENT; TORGUSON; HOLDER, 2009; SLOCUM et al., 2009). 

The HSV additive color model is a three-dimensional conical model, and its main 

components are hue, saturation, and value (DENT; TORGUSON; HOLDER, 2009; 

PETERSON, 2014). The hue corresponds to the circle of the cone with angular variation from 

0 to 360°, in which the main colors (red, yellow, green, cyan, blue, and magenta) are set each 

60° (DENT; TORGUSON; HOLDER, 2009; SLOCUM et al., 2009). 

The value corresponds to the vertical axis of the cone, ranging from 0 to 100%, 

from black at the bottom to white on the top. Saturation corresponds to the horizontal axis, 

ranging from 0 to 100%, varying from the center to the edge of the cone, from white in the 

center to the pure color in the full circle of the cone when the value is 100% (DENT; 

TORGUSON; HOLDER, 2009; PETERSON, 2014). 

 

2.5.3. Color contrast and differentiation 

 



32 
 

  

As previously mentioned, the ‘good’ use of colors contributes positively to visual 

search tasks (WILLIAMS, 1967). Robinson et al. (1995) affirmed that “the choice of colors is 

vital to the map’s success in portraying the geographic information effectively and, equally 

important, is not creating bothersome combinations and contrasts that are garish or draw 

unwanted attention.” In addition, the color’s visual variables (hue, saturation, and value) are 

instrumental in getting the perceptual grouping and figure-ground segregation (DENT; 

TORGUSON; HOLDER, 2009).  

However, working with colors is not always easy. When color is used, two visual 

phenomena may occur, the simultaneous contrast (ITTEN, 1970), also known as induction 

(BREWER, 1992), and the successive contrast (BECK, 1970), also known as differentiation 

(BREWER, 1992). Itten (1970) identified that simultaneous contrast or induction could be 

captured only by the eye (Figure 5). As pointed by Goldstein and Brockmole (2009), 

simultaneous contrast is “the effect that occurs when surrounding one color with others changes 

the appearance of the surrounded color.” 

 

Figure 5 – Example of simultaneous contrast or induction 

 
From: Goldstein and Brockmole (2009) 

 

The successive contrast or color differentiation is a phenomenon that occurs 

spontaneously when observing colors in different contexts (Figure 6). For example, a gray 

square looks darker when placed on a light background, and that same gray appears lighter 

when set on a dark background (BREWER, 1992). Although having complete control is not 



33 
 

  

possible, good map designs tend to minimize the occurrence of this phenomenon (DENT; 

TORGUSON; HOLDER, 2009). 

 

 

 

Figure 6 – Example of successive contrast or color differentiation 

 

From: Goldstein and Brockmole (2009) 

 

2.5.4. Subjective reactions to color 

 

People have color preferences, they give meaning to colors, and their behavior can 

be affected by colors (DENT; TORGUSON; HOLDER, 2009). Throughout life, the preference 

for colors changes. The study conducted by Helson and Lansford (1970) presents aspects about 

the color combination preference: the most pleasant color combinations result in a significant 

difference in value, which allows the segregation of figure-background; more excellent 

backgrounds are lighter or darker, intermediate values tend not to please; in general for figures 

the shades between green and blue or with little gray are pleasant, and shades in yellow, 

yellowish-green or with much gray are unpleasant; and mainly, the figure must be either lighter 

or darker to stand out from the background. 

Regarding the association of colors with emotions or sensations, Sharpe (1974) 

shows that: warm tones (red, orange, and yellow) tend to be associated with excitement, 



34 
 

  

stimulation, and aggressiveness; blue and green tones give the feeling of security, calm, and 

peace; shadows of brown, gray and black melancholy, depression and sadness. 

On using colors in maps, Robinson (1967) emphasized that colors should be used 

as tools to simplify the map and transmit information more clearly. Colors are used as tools to 

achieve the figure and the background of the map. The colors affect the perceptibility of the 

map. That is, legibility and visual acuity will be affected by the result of the use of colors. The 

colors, especially the hue, cause subjective reactions and sensations in people, which must be 

considered when planning the map. 

 

2.6. Map evaluation 

 

Map evaluation has been used to measure the quality of the cartographic 

communication by considering specific user tasks, focused on efficacy (ex.: number of correct 

answers), efficiency (ex.: response time), and preference (ex.: level of satisfaction). In addition 

to quantitative assessments, it is possible to use qualitative map assessments. Suchan and 

Brewer (2000) present qualitative methods for map evaluation, focusing on three areas: verbal 

data from questionnaires, interviews, and think-aloud protocols; methods of ethnographic data 

analysis; and analysis of documents such as maps, images, or texts. 

Questionnaires can have open and/or closed questions, depending on how 

structured the questionnaire is (SUCHAN; BREWER, 2000). Scales can be applied to evaluate 

questions or statements, such as Linkert-Scale, or NASA-TLX (SUCHAN; BREWER, 2000; 

GHUNTER, 2003; PRESTON, 2009; PUGLIESI et al., 2013). “Interviews are related to 

questionnaires but occur as social interaction. For instance, in focused interviews, participants 

discuss their experience of a situation created by the researcher” (SUCHAN; BREWER, 2000). 

And, as pointed out by Young and Stanton, “if you want to find out what a person thinks of a 

device, you simply ask him/her” (YOUNG; STANTON, 2005).  



35 
 

  

The interview technique which can be used for evaluation has advantages because 

questionnaires are familiar for most people, and in some cases, it can be applied via the internet. 

However, there are disadvantages, such as evaluation analysis can take a considerable time, and 

sometimes the protocol used to gather information from the interview can induce misleading 

results (YOUNG; STANTON, 2005). 

A relevant form of interview is semi-structured. According to Stanton (2005), “the 

semi-structured approach is recommended for a product evaluation interview.”  A semi-

structured interview follows an agenda of questions, but it is not random (completely 

unstructured) or strict (completely structured) (STANTON, 2005). The idea of having an 

agenda of items allows to extract relevant information from unfamiliar users, for example, 

"Let's talk about visual clarity: Did you think the information is clear and well organized?" 

(YOUNG; STANTON, 2005).  

  



36 
 

  

3. Cartographic design for the topographic maps 

 

The method employed to design the maps is depicted in a workflow presented in 

Figure 7. The following subsections present the adopted steps. The workflow used to construct 

the topographic maps, such as a definition for the interdependent variables and data selection, 

the preparation of a cartographic database, and the construction of a cartographic design. 

 

Figure 7 – Workflow for creating the three designs 

 
Font: The Author (2021) 

 

 

 

 

 



37 
 

  

3.1. Planning the map 

 

In the context of cartographic design, Decanini and Imai (2000) used the term 

‘interdependent variables’ as a group of aspects related to the stage of planning a thematic map: 

purpose, user and task, study area, scale, information, projection system, reference system, 

among others. Tyner (2010) and Muehlenhaus (2013) point out some similar aspects for 

interdependent variables as the map purpose. The topographic map is a product constructed for 

general purposes (ROBINSON et al., 1995) and it can be used for inventory mapping, town 

plans, orientation, navigation, etc. (SSC, 2005). According to the Swiss Society of Cartography 

(2005), topographic maps at 1:10,000 scale are mainly used for local applications in 

administration, business, planning, and data analysis.  

Since no cartographic database was not found to have building features available 

for the western region of the São Paulo state, it was necessary to use a cartographic database 

available at GeoSampa, a geo-portal of the São Paulo city government. GeoSampa provides 

several types of spatial data of the São Paulo city for free, including blocks, streets, vegetation, 

and buildings in GIS format (Shapefile ESRI). After choosing São Paulo city to use in this 

research, a refinement was performed, in which some neighborhoods were selected to be used 

as areas of study, at 1:10.000 scale (Figure 8). The reference system SIRGAS2000 was adopted, 

as well as Universal Transverse Mercator projection, for a database located in zone 23, southern 

hemisphere.  

 

 

 

 

 

 



38 
 

  

 

Figure 8 – Area of study in the São Paulo County highlighted by black contour selected for this 

research 

 
From: The author (2021)  

 

3.2.  Preparing the cartographic database 

 

A cartographic database was prepared as a framework to design the maps symbols. 

ArcGIS 10.8 was used to organize the database, including new layers in Shapefile ESRI format, 

which were opened in Quantum GIS 3.6.3 (QGIS) to design the cartographic representations. 

Since a street layer was not available in an area spatial dimension, it was necessary to construct 

one. The feature classes chosen to be used in this work are shown in Table 01. As it can be seen, 

it was not added contour line, spot height, water course and body, landmark, and so on. The 

low number of feature classes chosen to be present on the map face followed the Gestalt law of 

simplicity, once the purpose of the participants was to compare different designs conceived for 

the buildings. 

 



39 
 

  

 

 

Table 1 – Specification of the feature classes selected for this work 

Category Class 
Spatial 

Dimension 

Measurement 

Level 

Source 

Construction 

Building Area Single class GeoSampa 

Block Area Single class GeoSampa 

Transportation 

System 

Street Line Ordinal GeoSampa 

Street Area Single class Author 

Vegetation 

Public Park Area Single class GeoSampa 

Green area Area Single class GeoSampa 

Single class = Qualitative data having just one type. 

From: The author (2021)  

 

The street feature class was labeled and classified according to the Código de 

Trânsito Brasileiro – CTB (Brazilian Traffic Code). That classification is used mainly in studies 

related to urban planning and hierarchizes streets in three levels of importance depending on 

the traffic speed: local (30 Km/h), collector (40 Km/h), and arterial (60 Km/h). Street data 

provided by GeoSampa with line geometry was used for labeling street names. An Erase 

operation was used in ArcGIS by subtracting blocks from the chosen area to create streets with 

area dimensions. The proposed representations for topographic maps were created in QGIS 

3.6.3. 

 

3.3.  Designing cartographic symbols for the proposed topographic maps 

 

Three approaches were considered to construct the proposals for the topographic 

map symbols for the context of the west region of the São Paulo state, at 1:10,000 scale. The 

first approach is visually coherent with the national cartographic conventions, which was 



40 
 

  

inspired by the proposal presented by Sluter et al. (2018), for the Paraná state. The second 

approach is visually coherent with the urban landscape of the western portion of the São Paulo 

state, which was inspired by the works of Kent (2005), Kent and Vujakovic (2009), Kent 

(2013), in which they brought reflections about the relationship between the topographic map 

aesthetics and the local landscape. The inspiration was also based on the findings presented by 

Raposo and Brewer (2014) on what they found out which landscape influences the topographic 

map preference when using orthoimage as background. The third approach is visually coherent 

with the navigational maps used frequently, which indirectly derive from the Gestalt laws 

(ELLIS, 1955; MACEACHREN, 1995) and the preference for past habits or experiences 

(DEARDEN, 1984; HAMMITT; KNAUF; NOE, 1989). The HSV model was used to set up all 

three designs. 

To compare the three proposals that may represent buildings, the same kind of street 

design was used. It was decided to work with the concept of visual association, such as Sluter 

et al. (2018), in which gray was employed to represent the paved street. The gray color is also 

used in the context of national spatial data infrastructure by BRASIL (2018) to depict streets 

and their topological relationship with other geographic features, in the matter of acquiring 

geospatial vector data. To achieve the Gestalt laws of closure and good continuation (figure-

ground), as well as contour3, (grouping), three different types of relationships were considered 

for blocks and streets: among different blocks, among different streets, and between blocks and 

streets. In those cases, both feature classes were outlined/contoured equally. The good 

continuation of the streets is also provided by the natural characteristic of the road layout. The 

gray color allowed to achieve perceptual grouping for the streets, which also provided 

segregation from the other features. 

 
3 The term “contour” used in Gestalt to refer to a buffer resource around an object (outline) is the same term used 

in Topography to refer to lines of the same altitude 



41 
 

  

The street names had the same design for the three propositions. The typeface 

established by the manual T34-700, the Sans Serif Narrow (SSNR), did not present legibility 

on the display. Then, for arterials and collectors it was decided to use Calibri typeface in dark 

gray color (HSV = 0,0,15; RGB = 38,38,38). Arterials were sized to 8 points, bold, countered 

by a white mask of 1.5 points. Collectors were sized to 7.5 points, regular, countered by a white 

mask of 1.5 points. Bold and regular styles were used as visual variables to represent names for 

arterials and collectors, respectively. The local streets names were omitted in favor of legibility. 

Intending to reduce the possible differences when the participants of the experiment were using 

different displays via the internet, two notebooks were used during the stage of design, an Acer 

Aspire ES1-572 n16c1 series, 15.6", LCD screen, and a Dell Inspiron 7000 series, 15", LED 

screen.  

 

3.3.1.  Visual coherence between the proposition and the national cartographic 

convention 

 

The national cartographic conventions present in the Technical Manual T34-700 

2nd Part must be used to represent geographic features on official topographic maps (BRASIL, 

2000). According to the manual, those specifications must be employed in the ‘final drawing’, 

that is, the map representation. The present study considers only a few features: buildings, 

streets, blocks, and vegetation, from several items presented in the manual. The specifications 

defined in the manual T34-700 for those features are shown in Table 02. 

An adaptation of the official conventions was achieved by using a dropper tool to 

get the feature color shown in the T34-700 since there is no color model associated with the 

features. The color employed for blocks was adjusted to represent the building; the hue was not 

changed in the HSV model, but the value was increased, and the saturation was decreased. All 

features of the same class for buildings were filled and outlined with similar colors to achieve 



42 
 

  

a high level of perceptual grouping. Darker the outline was an option to reach figure-ground 

segregation by contour, a Gestalt law. Considering the blocks, to avoid using a reddish magenta 

for the whole space, a soft gray tone was used to differentiate them from the buildings and the 

‘empty spaces’ in white areas. The dropper tool was also used to get the graphical characteristics 

of the arboreal natural vegetation, while the hue was maintained, the value and the saturation 

were decreased. 

Table 2 – Graphical specification of some feature classes presented in the T34-700 manual 

Category Class Spatial Dimension Symbology 

Constructions 

Buildable area, 

represented by 

scale (page 49) 

Area 

 

Generic 

Building 

(page 27) 

Area 

 

Transportation 

System 

Street (page 2) Line 

 

Paved road 

(page 1) 
Line 

 

Landmark 

Natural 

arboreal 

vegetation 

(page 53) 

Area 

 

From: Adapted from T34-700 (BRASIL, 2000)  



43 
 

  

 

3.3.2. Visual coherence between the proposition and the western landscape in the 

state of São Paulo 

 

According to Kent (2005), the beauty of a topographic map can be achieved by the 

landscape represented on the map face. Constructing a topographic map representation 

according to the landscape character was inspired by the works of Kent (2005), Kent and 

Vujakovic (2009), and Kent (2013), in which they formulated ideas about the aesthetics of the 

topographic map and the local landscape. Also, another inspiration came from the findings 

provided by Raposo and Brewer (2014), which found the preference for the map design was 

mainly influenced by the represented landscape. They used orthoimage as background for the 

topographic maps and applied several map designs for the whole U.S. country. Thus many 

landscapes and designs were assessed. 

Concerning the landscape preference, the findings provided by Dearden (1984) 

indicated that the participants’ choice was influenced by some factors, among them, familiarity 

with the observed landscape and the place where the person has lived in the last five years. 

Dearden (1984) used photographs taken from the observer's perspective. However, the 

Brazilian official topographic maps are projected orthogonally, and according to Santos (1988), 

the perspective and the observer's knowledge could influence the perception of the landscape. 

It was necessary to identify some characteristics of urban landscapes to represent a 

topographic map considering similarity with the landscape of some municipalities in the São 

Paulo state. Satellite images, available for free of charge were obtained from Google Earth and 

orthoimages at 1:10,000 scale, available from IGCSP were clipped in true colored composition. 

In Figure 9, clippings from large and small cities located in the São Paulo state are presented. 

The most distinctive visual patterns of the landscape seen from the orthogonal 

views were identified between the metropolitan region located in the capital (Figure 9: c, f) and 



44 
 

  

the countryside cities by the west and east (Figure 9: a, b, d, e, g, h, j, k) and the east coast 

(Figure 9: i, l). The landscape located in the metropolitan region is remarkably shaded by gray 

shades being predominantly compound by high buildings, shadows of the buildings, and paved 

urban roads (Figure 9: c, f; Figure 10). The landscape located in the west and east countryside 

cities are remarkably shaded by orange, brown, green and gray shades (Figure 9: a, b, d, e, g, 

h, j, k; Figure 11). 

 

Figure 9 – Clippings of satellite images in the true color composition of cities in the São Paulo state 

 

 
From: Google Earth (2021) 

 

 



45 
 

  

Figure 10 – Clippings of satellite images in true color composition from the capital and metropolitan 

region of the São Paulo state 

  
From: Google Earth (2021) 

 

The small and medium cities located in the western countryside present roofs 

predominantly in shades of orange, red and brown, followed by shades of gray, similar to 

pavements made of cement or concrete (Figure 11). In that sense, the representation of blocks, 

buildings, and urban roads were based on those color compositions identified by satellite images 

(Figure 11 a, b, c, d, e, f) and orthoimages at 1:10,000 scale (Figure 11 g, h, i, j, k, l).  

A dropper tool was used to get the color in the RGB model from the roofs of a few 

cities from the west of São Paulo state. The colors were converted to the HSV model. Hue, 

saturation, and value were adjusted empirically to achieve a pleasant shade. It was obtained an 

orange color, which was used to fill and contour features identified as buildings to use with the 

cartographic database of São Paulo city. 

 

 

 

 

 

 



46 
 

  

Figure 11 – Clippings from municipalities of the western of the São Paulo state: a, b, c, d, e, f satellite 

images in true color composition, 2020 g, h, i, j, k, l orthoimages at 1:10,000 scale, 2010 

 

 
From: a, b, c, d, e, f Google Earth (2021) and g, h, i, j, k, l IGC (2010) 

 

3.3.3. Visual coherence between the proposition and the digital navigational maps 

used frequently 

 

The proposition visually similar to navigation maps frequently used was inspired 

by the findings related to the landscape preference provided by Dearden (1984). Personal 

familiarity with the landscape was the determining factor. Hammitt, Knauf, and Noe (1989) 



47 
 

  

identified that the previous experience in a recreational activity influences their preference for 

carrying it out. In addition, the Gestalt law of experience or habit is a factor that influences the 

recognition of visual patterns (ELLIS, 1955; MACEACHREN, 1995). Two navigational maps 

that could be used frequently worldwide, Google Maps and Waze, were considered as 

inspirations. Those two maps are represented in white and gray shades (Figure 12). Then, the 

gray coloration was chosen. 

 

Figure 12 – Clippings of digital navigational maps: Google Maps above and Waze online bellow 

 
 

 
From: above, Google Maps (2021), and bellow, Waze online (2021) 

 

Designing a map in gray shades is a task of great complexity. Because only the 

color value should vary for buildings, blocks, and streets, while hue and saturation stay the 



48 
 

  

same in the HSV model. The grayscale makes it challenging to perform the perceptual grouping 

between features classes by levels of gray, but mainly the figure-ground segregation. Therefore, 

careful use of the visual variable value was required to not compromise the legibility and 

differentiation of features. The advantage of a map in shades of gray could be the flexibility for 

the later addition of thematic layers, in which the neutrality of the map could not influence the 

perception of the themes. Muehlenhaus (2013) and Peterson (2014) indicate the use of gray as 

an option for the background of web maps. 

 

3.4.  Proposed designs and graphical specifications 

 

A graphical specification was defined to be replicated in future works (Table 03). 

As previously mentioned, it was decided to keep the number of features on the map face, using 

a Gestalt concept called simplicity, by omitting contour line, spot height, watercourses and 

water bodies, landmarks, and so on. The reason for that consists of focusing on the reading of 

features that need to be considered during the evaluation.  

 

Table 3 – Graphical specification of the cartographic symbols 

Category Class 
HSV RGB Graphic 

Symbol 
Width 

Fill Contour Fill Contour 

Constructions 

Buildings 

Pink 
0,10,100 0,30,90 

255,230,

230 

230,161,

161 

 
0.11mm 

Buildings 

Orange 

15,15, 

100 
15,40,90 

255,227,

217 

230,161,

138 

 
0.11mm 

Buildings 

Gray 
0,0,90 0,0,70 

230,230,

230 

179,179,

179 

 
0.13mm 

Blocks 0,0,95 0,0,20 
242,242,

242 
51,51,51 

 
0.10mm 

Transportation 

System 
Street 0,0,80 0,0,20 

204,204,

204 
51,51,51 

 
0.10mm 

Vegetation 

Public 

Park /  

Green 

Area 

130,10, 

90 
- 

207,230,

211 
- 

 

- 

From: The author (2021) 

 



49 
 

  

The graphical specification presented in Table 03 was used to create the proposed 

maps. Figure 13 shows clippings of each design that was sized according to the actual size 

which was designed for the 1:10,000 scale. All the maps designed are shown in the Figures 

from 14 to 20. Each design was named according to its visual coherence: Pink for national 

cartographic conventions, Orange for the western landscape of the state of São Paulo, and Gray 

for frequently used navigational maps. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



50 
 

  

Figure 13 – Clippings of proposed designs applied for the area 01 at 160 dots per inch resolution, 

zoom 100%: Pink (on the top), Orange (in the middle), Gray (at the bottom) 

 

 

 
From: The author (2021)  

 



51 
 

  

The nine maps shown in Figures 14 to 16 were used during the qualitative 

evaluation stage (for further information, see chapter 4). 

 

Figure 14 – Proposed designs applied for the area 01: Pink (on the top), Orange (in the middle), Gray 

(at the bottom) 

 
 

 
 

 
From: The author (2021)  



52 
 

  

Figure 15 – Proposed designs applied for the area 02: Pink (on the top), Orange (in the middle), Gray 

(at the bottom) 

 
 

 
 

 
From: The author (2021)  

 

 

 



53 
 

  

Figure 16 – Proposed designs applied for the area 03: Pink (on the top), Orange (in the middle), Gray 

(at the bottom) 

 
 

 
 

 
From: The author (2021)  

 

The sixteen maps shown in Figures 18 to 20 were used during the quantitative 

assessment stage (for further information, see chapter 4). 

 



54 
 

  

Figure 17 – Proposed designs applied for the area 04: Pink (on the top), Orange (in the middle), Gray 

(at the bottom) 

 
 

 
 

 
From: The author (2021)  

  

 

 



55 
 

  

Figure 18 – Proposed designs applied for the area 05: Pink (on the top), Orange (in the middle), Gray 

(at the bottom) 

 
 

 
 

 
From: The author (2021)  

 

 

 



56 
 

  

Figure 19 – Proposed designs applied for the area 06: Pink (on the top), Orange (in the middle), Gray 

(at the bottom) 

 
 

 
 

 
From: The author (2021)  

  

 

 



57 
 

  

Figure 20 – Proposed designs applied for the area 07: Pink (on the top), Orange (in the middle), Gray 

(at the bottom) 

 
 

 
 

 
From: The author (2021)  

 

  



58 
 

  

4. Assessment of the user’s preference 

 

The workflow that considers the processes of evaluating the cartographic 

representations is presented in Figure 21. The following subsections present the steps developed 

to assess the user’s preference for the proposed topographic map symbols, such as the 

documents for appreciation of the Local Research and Ethics Committee, the characterization 

of the subjects, the experimental design, and the procedure for applying the experiment. 

 

Figure 21 – Workflow of the steps used to evaluate the user’s preference for the proposed designs  

 
From: The Author (2021)  

 

 



59 
 

  

4.1. Preparing the questionnaire and commitment documents 

 

It was necessary to provide to the Local Research and Ethics Committee a set of 

documents that emphasize the physical safety and moral integrity of the participants and the 

confidentiality of the data to evaluate the quality of the cartographic communication through 

the preference that involves an assessment with people. It was necessary to complete three 

documents and prepare two more items to accomplish the interviews. 

These three documents consisted of the Termo de Compromisso – Legislação 

(Commitment Term – Legislation) to be signed by the researcher, the Termo de 

responsabilidade e compromisso para uso, guarda e divulgação de dados e arquivos de 

pesquisa (Term of Responsibility and Commitment to use, save and publish data and files) to 

be signed by the researcher, and the Termo de Consentimento Livre e Esclarecido (Free and 

Informed Consent Term) to be assigned by the researcher and the participants. Those three 

documents were made available digitally to each participant. 

The procedure of the interview was detailed in a script that describes the scenario 

with instructions about how to perform the tasks, as well as a basic experimental test design. 

The script and the test design were inspired mainly by Raposo and Brewer (2014) and Deeb et 

al. (2015), whose focus is on determining the participants' preference about specific 

characteristics of the map design. The documentation is found in Appendix A. 

 

4.2.  Subjects 

 

The experiment to evaluate the user’s preference about the proposed maps during 

the pandemic caused by SARS-CoV-2 was conducted through the internet. As requirements, 

participants should live in the western region of the state of São Paulo for at least five years, as 

considered by previous literature (DEARDEN, 1984). Also, the background was considered 



60 
 

  

since it could influence the landscape perception (SANTOS, 1988). Two groups were formed, 

each one containing 25 participants, totaling 50. One group called architects was compound of 

subjects majored in Architecture and Urbanism, as well students enrolled in last year of the 

same type of undergraduate course. The group compound by architects had 3 males and 22 

females (mean age 27.92; SD = 9.54; range 21 to 52). The other group called non-architects 

was compound of subjects majored in other areas (Cartographic Engineering, Civil 

Engineering, Environmental Engineering, Computer Sciences, and Mathematics), or enrolled 

in last year of Cartographic Engineering undergraduate course, or postgraduate students in 

Cartographic Sciences, and teachers of cartography disciplines as well. The non-architects 

group had 12 males and 13 females (mean age 32.44; SD = 12.16; range 23 to 64). All 

participants declared having normal vision color and being map users. 

 

4.3.  Experimental design 

 

The experimental design for the quantitative and ordinal data was a 3 x 2 mixed 

factor to compare design (visual coherence with national cartographic conventions, the urban 

landscape of western São Paulo, navigation maps used frequently; within-subject), background 

knowledge (architects and non-architects; between-subjects). The experimental design focused 

on identifying the order of preference among the map designs, mainly if there would be any 

more preferred than others by the participants. The dependent variables were based on 

subjective measures: level of satisfaction for each design and order of preference among the 

designs. 

 

 

 



61 
 

  

4.4.  Procedure 

 

The experiment consisted of interviewing each participant individually through a 

video conference on the Google Meet platform. Some days before running the test, the 

experimenter invited the subjects to participate. The subjects were instructed about ethical and 

legal documentation, and the documents were provided to them.  

On the day scheduled for the interview, the participant accessed Google Meet, and 

they were instructed about the experiment. Before starting the assessment, the subject was 

advised that the videoconference would be recorded. After the consent, the participant informed 

a set of individual characteristics that could be used for further analysis, if necessary: name, 

age, gender, whether they have a normal color vision or not, and background knowledge in 

Architecture or other areas. The experimenter provided the web links to access the maps, and 

he instructed the participant about how to use the website.  The website4 was built by using the 

Google Platform Blogger tool, which allows viewing images with high resolution, the same as 

produced. Each blog post presented a set of three maps to see one above the other, then 

separately by clicking on each map. That strategy facilitated the access by the subjects because 

each page had a unique and direct access link.  

Although the procedure of reading maps through a website facilitated the access in 

a critical moment of restrictions due to the pandemic caused by Sars-CoV-2 in Brazil, that 

choice implied in the absence of rigorous laboratory control of the experiment (e.g., type, 

resolution, and size of the display). The unique restriction was that the device should be a 

notebook or computer, avoiding tablets and smartphones, whose displays would be significantly 

different. Each participant accessed the website with their device. Applying interviews remotely 

by videoconferences may have brought the map access closer to the real usage of an online 

topographic map. 

 
4 https://gustavograssi.blogspot.com/ 



62 
 

  

The experiment procedure is presented in Figure 22. In the experiment introduction, 

each participant observed a total of three orthoimages which IGCSP produced at the 1:10.000 

scale (see Appendix A => Roteiro e questionário do experimento => Primeira Etapa). 

Important to remember is that the area shown through the orthoimages to the participants does 

not correspond to the regions shown through the maps, with the reason of not having an 

available cartographic database of buildings for the west portion of the São Paulo state. Each 

image exhibits a city seen from the top, in which they are located in the western part of São 

Paulo: Presidente Prudente, Presidente Venceslau, Dracena. 

An introduction of the experiment was accomplished by showing the geographic 

context. After that, three maps were presented corresponding to each design proposed 

separately, so the participant could judge what they think about the designs, stimulated by these 

questions5:  "What do you think about the symbol designs of the presented map?", "What do 

you think about the representation of the streets?", "What do you think about the representation 

of the blocks?", "What do you think about the representation of the buildings?", "What did you 

most like on the map? Or what stood out positively?", "Is there something on this map that 

annoyed you or made you feel uncomfortable? Or that stood out negatively?".  

In the sequence, the presentation of the three designs allowed the participant to 

assign a score and indicate an order of her or him preference by answering these questions6: 

"Please, assign a score from 0 to 10 for each of the designs proposed for topographic maps, 

where 0 is the lowest score and 10 the highest, taking into account the readability of streets, 

blocks, buildings, vegetation, and street names.", "Please, indicate the order of preference for 

the designs proposed for these topographic maps." That procedure was performed four times, 

each one for a different urban area. The reason for that was to check the average of the design 

scores. The experiment lasted approximately 25 minutes.  

 
5 The experiment was performed in native language, Portuguese-Brazil, thus all questions were translated and 

may not correspond exactly to the literal original expressions. 
6The experiment was performed in native language, Portuguese-Brazil, thus all questions were translated and 

may not correspond exactly to the literal original expressions. 



63 
 

  

 

Figure 22 – Steps used during the interview with the questionnaires 

 
From: The author (2021) 

 

 

 

  



64 
 

  

5. Results 

 

The data obtained from the interviews with the subjects were analyzed in three 

sequential stages: opinions regarding the designs, rating score given to the designs, and order 

of preference by the designs.  

 

5.1.  Analysis of the opinions regarding the characteristics of the designs 

 

The analysis of the patterns that can briefly describe the designs was accomplished 

by searching for repeated terms through the opinions provided by the subjects. Data were 

organized into different tables according to the group of subjects (architects, non-architects) 

and the type of design (Pink, Orange, Gray). The responses in the Portuguese language are 

found in Appendix B. The subjects’ opinions were categorized into positive and negative 

considerations. Some suggestions for improving the characteristics of the designs are also 

present, in which they can provide insights that could serve as recommendations for future 

works. 

The responses provided by each group of subjects were organized in three different 

ways: the type of design, general opinions about the map, and specific opinions about the 

cartographic symbols. The opinions7 provided by the group of architects can be found in Tables 

04 and 05 (Pink), Tables 06 and 07 (Orange), and Tables 08 and 09 (Gray). For the non-

architects group, the opinions can be found in Tables 10 and 11 (Pink), Tables 12 and 13 

(Orange), and Tables 14 and 15 (Gray). 

Regarding the general opinions, the responses provided by the architects and non-

architects were similar. The most important general opinions pointed out by both groups reveal 

 
7 The content of all qualitative responses in Tables 04 to 15 were performed in native language, Portuguese -

Brazil, thus their translation may not correspond to the literal original expressions.   



65 
 

  

the Orange design being the best approach to represent contrast and the Gray design being the 

worst case. Some subjects who declared those maps in Orange having the best contrast affirmed 

that it was possible to see the buildings at first sight, while those maps with the worst contrast 

required more time or effort while reading it. In addition, both groups of participants affirmed 

that they had difficulty seeing the buildings in Gray and Pink designs, a factor not reported in 

the Orange. 

Some subjects presented suggestions to improve the legibility of the designs. For 

example, one of the architects pointed out a concern with the map being used by visually 

impaired people, indicating that all designs were characterized by tones similar to pastels. 

Considering the Gray design, 45% of the architects recommended increasing the contrast 

between different cartographic symbols to see urban fills and voids. Considering non-architects, 

none of them pointed out general recommendations for the Orange design. About 30% 

suggested increasing color saturation in the Pink design, and around 15% mentioned increasing 

the color contrast in the Gray design. 

Each cartographic symbol was also judged individually. Some participants from 

both groups pointed out the green color used for vegetation as easy to see but unattractive. For 

example, about 20% of the architects and 10% of the non-architects suggested darkening the 

green color to improve the contrast. 

 

5.1.1. General analysis by the design composition 

 

The Pink design was quite well evaluated by architects and non-architects (Tables 

04 and 10). About 70% of the architects and 40% of the non-architects observed pink as 

appropriate to differentiate the map features. Around 50% of the architects and 70% of the non-

architects judged that seeing the map features was comfortable and pleasant. About 50% of the 



66 
 

  

architects and 35% of the non-architects praised the use of soft colors. On the other hand, about 

20% of the participants criticized colors as ‘too soft’, indicating a certain lack of contrast. 

The Orange design was also quite well evaluated by both groups (Tables 06 and 

12). About 50% of the architects and 65% of the non-architects indicated that differentiating 

the map features was easy. About 70% of both groups judged the design as comfortable or very 

comfortable. Around 30% of the architects and 20% of the non-architects praised the colors 

chosen for the map. However, for 20% of the architects and 12% of the non-architects, the 

Orange was slightly inconvenient. 

The Gray design was pretty criticized negatively by both groups (Tables 08 and 

14). Just a minority of the subjects presented positive opinions. About 20% of the participants 

in each group affirmed that the reading was comfortable and enjoyable. Around 25% of the 

architects indicated that a grayed-out map could help insert thematic layers. On the other hand, 

approximately 85% of the architects and 70% of the non-architects pointed out a lack of contrast 

and difficulty differentiating the map features. Additionally, about 40% of the architects and 

30% of the non-architects indicated this design as more time-consuming to accomplish the 

reading. In addition, around 45% of the architects and 40% of the non-architects pointed out 

that the design was 'very gray,' ‘monochromatic,’ ‘boring.’ 

 

5.1.2. Analysis by toponymy and individual cartographic symbols 

 

Street labels were scored positively on Pink and Orange designs. Around 65% of 

the architects (Tables 05, 07, and 09) and more than 85% of the non-architects (Tables 11, 13, 

and 15) affirmed they could be read easily. Considering the answers provided by design, nearly 

50% of the participants from both groups reported Pink as outstanding. About 30% of the 

architects and 60% of the non-architects reported Orange design with good readability. 



67 
 

  

Approximately 25% of the architects and 12% of the non-architects reported the Gray design 

as readable. 

Regarding the Pink design, the architects judged the buildings positively and 

negatively almost in the same proportion (Table 05). About 25% of the non-architects stated 

the pink color for building as readable, and around 20% reported that pink was uncomfortable. 

In the matter of suggestions, 20% of the architects recommended darkening the buildings. 

Considering non-architects (Table 11), nearly 50% of the respondents indicated difficulty in 

seeing the outline of the buildings, while 30% reported the color as uncomfortable. Regarding 

suggestions, 20% of the non-architects who participated in this study recommended increasing 

or improving the outline. 

Considering the Orange design, about 55% of the architects pointed out that the 

buildings were easy to see and an excellent color to differentiate between urban fills and voids. 

Around 10% criticized the darker colors, and approximately 10% suggested darkening the 

outline (Table 07). About 60% of non-architects (Table 13) indicated that the Orange facilitated 

to see the area and its outline, and 45% said that the Orange color was the best choice for the 

buildings. On the other hand, 20% of non-architects indicated difficulty in seeing the outline. 

Considering the Gray design (Tables 09 and 15), only about 15% of both groups 

stated that the gray representation for the buildings was good. On the other hand, 70% of the 

architects and 20% of the non-architects reported difficulty reading the buildings. In addition, 

60% of the architects and 40% of the non-architects pointed out that urban voids and urban 

spaces were confusing. Additionally, 70% of the non-architects indicated a lack of contrast, 

45% of the architects and 12% of the non-architects recommended improving the outline. 

 

 

 

 



68 
 

  

 

Table 4 – Architects' opinions about the Pink design 

Positive feedback N° Negative feedback N° 

Easy to identify all features / Good 

readability / clear / Does not cause 

confusion / very clear 

17 Difficult to identify voids, green areas, and 

buildings 

5 

Visually smooth / light / pleasant light 

tones / soft colors 

13 Absence of contour lines / topography 4 

Comfortable / visually comfortable / 

aesthetically pleasing  

12 Lack of legend / map elements 4 

Easy to understand the meaning of the 

features / easy to read / communicative 

8 Poor to add hatches/themes 1 

Map is good / very good / well done / it 

looks good 

8 Poor to find spaces 1 

I really like it / very cool/cool 6 Blurred vision when looking a long time 1 

More comfortable / more enjoyable / more 

preferred 

4 I do not like pink 1 

It is not visually tiring / it does not disturb 

the vision 

3 Pink and green do not look comfortable 

together 

1 

Interesting 2 It demands a lot of attention identifying details 1 

Good map for architects because of the 

presence of buildings, green areas 

1 Overload the vision a little 1 

Organized map 1 Not very detailed 1 

From: The author (2021) 

 

Table 5 – Architects' opinions by feature class in the Pink design 

Category Positive feedback N° Negative feedback N° Suggestions N° 

Buildings 

Easy to view buildings / 

communicate well the 

urban voids 

6 Buildings cause a little 

discomfort 

5 Darken buildings more 5 

Good ability to check 

volume by geometry 

 

3 Poor to view buildings 

in dense areas 

5 Improve/increase 

building outline 

3 

  Reddish pink tends to 

be associated with 

danger, severity, etc. 

1 Maybe use another color 

for buildings 

1 

  Buildings lack contrast 1  

  Pink looks like a blur 1 

Streets 

Highlights / well-

defined  

12 There is not much 

difference between the 

categories of streets 

1 Give more emphasis / 

highlight more the 

classification of the 

street 

2 

Road classification well 

perceptive 

1 

Label 

Legible/easy to read 

labels 

23 The absence of labels 

for local streets can 

make it difficult for 

dummies to read 

1  

  Labels are not readable 1 

Green 

areas 

Good perception 2 Confusion of green 

with blue 

1 Darken even more / 

increase contrast 

4 

Voids / 

Blank 

areas 

Clearly visible  1   

Blocks Well defined 3     

From: The author (2021) 

 



69 
 

  

 

Table 6 – Architects' opinions about the Orange design 

Positive feedback N° Negative feedback N° 

More visually pleasing / more comfortable / 
Comfortable / visually comfortable 

17 Not so pleasant / little discomfort 5 

Good readability / clear / Easy to identify all 

features on the map 

12 Absence of contour lines/topography 3 

I like it / very beautiful/interesting 8 Most visually loaded 2 

Stronger / darker / more attractive color 7 Less comfortable 2 

Good contrast 7 Poor to add hatches/themes 1 

More alive / more vibrant / more colorful / more 

organic / brings joy 

5 Two [streets and buildings] competing for 

graphic weights 

1 

Does not cause confusion / discomfort / 

tiredness  

4 Scatter information 1 

Visually smooth / does not affect overloading 

vision 

3 There is no excellent difference between the 

features 

1 

The map is good / well done 2 Orange changes color perception 1 

Differentiation of features at first sight 2 Adding layers could get a little heavy 1 

Good for general use 1 Absence of legend 1 

Better detailed 1 It takes longer to discern features 1 

Good map for architects due to the presence of 

buildings and green areas 

1 Less pleasant 1 

It causes a little confusion 1 

From: The author (2021) 

 

Table 7 – Architects' opinions by feature class in the Orange design 

Category Positive feedback N° Negative feedback N° Suggesti