Depth Perception in Daylight

(1)

Master Thesis

KTH - Architectural Lighting design & Human Health

Lighting Laboratory

royal institute of Technology KTH, campus Haninge course code: HS202X

Tristan rudebeck Haar

September 2015

an approach to depth perception through the illumination of diffuse daylight

Tutor: carlo Volf, MAA phD

(2)

Acknowledgment

I am very grateful for my tutor carlo Volf, for helping and guiding me through this thesis.

Sincere thanks to the in- spirational teaching staff at ktH lighting laboratory.

Also I would like to give thanks to my close friends and family: Helle, dit- mar, Ane, and Jette, for whom have supported me when I needed it the most.

(3)

Abstract

the focus of this thesis is to investigate if diffuse daylight affects human depth perception. It is built upon previous knowledge and methods of observ- ing and perceiving light brought into a research that experiment with different spatial contexts through scale models. the central position of perceptual cues within the human visual field is discussed in relation to perceptual depth and visual elements

the result of the performed experiement showed a possibility for diffuse daylight to have an effect on the perception of depth.

Having the knowledge of building with daylight will lead to a better understanding of how daylight is affecting our perception of spaces, which poten- tially can improve the ability of creating sustainable perceptual spatial experiences when designing and building with daylight as an integrated part of the design.

(4)

tabel of Contents

1.0 introduction 4

1.01 the Áim and ojectives 5

1.02 Boundaries 5

2.0 theoretical

background 7

2.1 daylighting 7

2.1.1 Introduction 7

2.1.2 daylight Variability 8

2.1.3 daylight Availability 8

2.1.4 daylight Health 10

2.2 Sun & Sky as Sources of light 12

2.2.1 Sky luminance 13

2.2.2 clear sky 15

2.2.3 overcast sky 15

2.2.4 the daylight Factor 15

2.3 Perception 19 2.4 Visual Space Perception 20

2.4.1 the textured ground 20

2.4.2 occlusion 22

2.4.3 context 23

2.4.4 linear Perspective 25

2.4.5 compression 25

2.4.6 Shading 26

2.5 light Perception 27

2.5.2 Brightness 28

2.5.3 Illumination 29

2.5.4 Visual gloom 29

2.5.5 light creates Space 30

2.6 depth Perception 34

2.5.6 Shadows 32

2.6.1 why do we Perceive depth? 34

2.6.2 depth from monocular cues 34

4.0 exPeriment 38

4.1.1 Inspiration and methodology 38

4.1.2 the Apparatus and materials 39

4.1.3 Procedure 40

4.1.4 location 40

5.0 reSult 41

5.1 Perceptual depth 41

5.2 Visual elements 42

(5)

6.0 diScuSSion 44

6.1 the Hypothesis 44

6.2 discussing Perceptual depth 44

6.3 discussing Visual depth elements 45

6.4 discussion Summary 46

6.5 experimental Problems 46

7.0 concluSion 47

referenceS 48

figureS 52

aPPendix a 53

aPPendix b 54

aPPendix c 55

(6)

6

the focus of the world today, is the consequences the human race has had on the global climate. Stand- ards and ratings for green building have been made to reduce the carbon footprint of buildings. this indi- cates the need for architecture to be sustainable. Use of recycled materials, positioning building according to the outdoor climate, costs and decomposition of the building when demolished are all important issues. the engineering too needs to be sustainable including plans for self-sufficiency, natural ventilation, and daylighting strategies.

Research in the advantages of using daylight as a potential light source has increased over the recent years and we are beginning to understand the ben- efits of being exposed to daylight on a daily basis.

Because we spend approximately 90% of our day indoor it is crucial to the design of the architecture how much daylight enters the indoor space. Including daylight into the planning of a space requires specific knowledge of how daylight affects the physical and psychological environment. daylight is dynamic and it has the ability to change how we perceive a space during a day and over a year. the dynamic changes of daylight during different timescales makes it difficult to implement in a static design. moreover, daylight can bring unwanted glare from daylight openings and the interior lighting design is thus forced to compromise between visual efficiency, visual comfort and aesthetic satisfaction, which means that the human eye needs to adapt to many different daylight scenes within the same space during the same day (Hopkin- son et al. 1966).

the complexity of designing with daylight is often seen in buildings with almost no window area or whole glass façades. The potential thermal and visual discomfort that is connected with large glass windows caused by the excessive sunlight penetrat- ing a space as well as glare caused by the absence of shadings can be avoided with proper daylighting solutions. the effect of the sun in a space with large daylight openings, are known and by tracking the sun it is possible to predict the amount of light and heat caused by the sun. contrary, when the sun is covered by clouds, it should no longer be a problem how much daylight is affecting a space because then main light source is the diffuse sky, which gives an even distribution of light. (Hopkinson et al. 1966).

when calculating and simulating daylight inside

buildings, the most commonly used standard is the c.I.e. Standard Sky luminance distribution. In the standard of the distribution, the c.I.e. overcast Sky standard gives the lowest possible sky luminance, which does not cause glare or any other discomfort issues. depending on the reflection from the outdoor environment and the accessibility of the view to the sky, a window will tend to be evenly illuminated by a diffuse sky distribution, and therefore an equal amount of light will fall into a space. the geometry of a space spreads the light according to the reflectance factor of the surfaces within the space and the space thus become visible to the eye and can be perceived. A visible space consists of depth, shape, colour and size, projected on to the retina in our eyes.

the retina is sensitive to colour, contrast, brightness, texture gradients and shadows (gibson 1950). It receives these inputs as a flat image, that has no depth in it, and from here the inputs are processed by the brain and perceived as a three-dimensional image (gibson 1950). But how do we perceive depth? what information in the retinal image creates the perception of depth?

It has been tested through experiments that uniform lighting works better for the central vision (liljefors & ejhed, 1990, p. 39 cited in lindh 2012, p. 22), and is one of the most used types of lighting distribution in working environments. moreover, it is noted in the cIe standards that the central vision needs a uniform lighting distribution to facilitate reading details, quoting: “In general, the more uniform the distribution of light in the visual field, the better one sees the visual task”, (m. Rea, 1993, pp. 98 cited in lindh 2012, p. 22). However, the uniform lighting of a space is not, according to lindh (2012), in favour for our spatial experiences and orientation abilities, but it is through the peripheral vision in which the spatial environment is perceived by the visible contrast from the light illuminating the vertical surfaces and through larger brightness contrasts. Furthermore, liljefors (1997, 2003) found that the distribution of light is crucial for our spatial experience, and from the assumption that we generally perform visual tasks better in spaces with uniform artificial light, it is interesting to discuss whether diffuse light causes any disturbance in our visual perception of a space.

therefore, my research question is: How can diffuse daylight create depth in a space?

1.0 introduCtion

“If a uniform, diffuse light is applied to a whole space it communicates

that no part or surface is worth emphasising or more important than any

other. A space like this is usually experienced as boring and insignificant

since it lacks the clearly defining contrasts, borders and gradients that the

gaze always searches for.” - O. Andersson 1988, p. 27

(7)

introduCtion

1.01 tHe AIm And oBJectIVeS

the aim of this thesis is to pursue how diffuse daylight affects the human spatial understanding of a space. more specifically, the objective is to assess how diffuse daylight affects the human ability to perceive depth, how spatial elements are experienced in a room with only light from diffuse daylight and how it creates an impression of the space as a whole.

the thesis is built on the assumption that diffuse daylight is affecting the spatial perception of depth.

It is my preliminary assumption from my experience as a lighting designer that the light we perceive is not the light that the eyes actually sees. that light in pictorial form is not able to construct an exact three dimensional space in the brain, but it is our perception which takes over our vision and creates a third dimension. gibson (1950) and S. Hasselgren (1969) described human perception, as being created by the visual world, which is stable and fixed, and we only perceive whatever we acquire in our visual field from moving around in the visual world. A lack of information in the visual field would then be fatal for what we experience. missing a cue for the perception of depth could then change our whole perception and understanding of the space.

1.02 BoUndARIeS

the aim of this thesis is to explore the visually perceived depth within the context of a space illuminated by diffuse daylight.

to justify what perception is about would require much more research within the science of psychology and physiology than this thesis can provide. It could be interesting to introduce stereoscopic vision

into this subject, because of our ability to see in three dimension is believed to be possible from retinal dis- parity. However, it is the interest and focus of this thesis to test the abilities of perceiving depth from the different components creating a space (wall, floor, ceiling and daylight openings) and the impression of the light that is reflected from the components to the back of the eye and seen on the retinal image (contrast, brightness, texture gradients and shadows).

this thesis will base its findings from the knowledge in the fields of architecture, environmental psychology and daylight studies. the main focus is to learn, to see lighting from a different perspective of how diffuse light is affecting a space through our perceptual sensitivity. this is an alternative to just making observation of how daylight affects already built spaces.

By not performing the experiments in residential living rooms, kitchens, and bedroom, the distraction of furniture and decoration, which could interfere with the perception of the space is avoided. An option for performing the experiments through computational daylight simulation software was tested during writ- ing this thesis, but the renderings did not produce a satisfying depth perception, even through the use of several computer-based daylight simulation and different rendering software. the experiment was chosen to be performed outside under the influence of real daylight and not within a daylight simulation laboratory, where the lower amount of light levels and distribution could have affected the depth perception in the pictures taken of the model. the final images of the testing is to be examined by the author for eliminating the chance of test participants having a background or experience within the lighting branch, which could create bias when comparing the test result.

Figure 1 | Vilhelm Hammershøi | Interiør, Frederiksberg Allé | 1900

(8)

Figure 2 | chiaroscuro | 2012

(9)

9 2.1 daylighting

2.1.1 IntRodUctIon

“We can easily forgive a child who is afraid of the dark; the real tragedy of life is when men are afraid of the light.” - Plato

that a civilized man adapts to his natural environment is a basic evolutionary feature of the 20th centuries thinking. daylight adaptation has been necessary for mankind, in his natural state, for evolving vision as a hunter and a predator. the idea of human vision being evolved or created to the adaptation of daylight implies that consistent exposure to artificial light creates a somewhat unpleasant adaptation. the indoor environment is often influenced only by artificial light sources, which makes the comfort level of a visual task significantly low because of the lack of satisfac- tory lighting conditions. the adequate comfort level of lighting for an indoor environment is based on the specifically visual task and it is therefore not possible to derive a simple method for good interior lighting.

the acceptable amount of light necessary for any visual task can be found from studies, however it is difficult to provide this amount from windows that are satisfying architecturally and without unwanted glare from the sky. the design of an interior lighting scenario is therefore a compromise between visual efficiency, visual comfort, and aesthetic. (Hopkinson et al. 1966)

“The Study of lighting derives from the ba- sic concept of adaptation, that is, the ability of the eye to adapt to the prevailing lighting condi- tions, and the whole body to adapt to the envi- ronment.” (Hopkinson et al. 1966, p. 1)

daylight is the most unique natural information for every living creature on our planet. daylight has from the beginning of mankind been informing the inhab- itants, initially about the difference between day and night. Architectural principles have followed daylight patterns, by means of windows and openings in walls of buildings in order to let light through to the indoor environment. daylight is therefore one of the main aspects for the development and definition of region- al architectural characteristic around the world. the window (orig. old norse ‘vindauga’) meant the winds eye has developed over the centuries from a hole in

the roof to the storytelling in christian churches and has always had the same purpose of letting in daylight and was until the twentieth century the primary source of lighting to all types of buildings. In the development of the electrical light source, daylight lost its position as the primary source and was suddenly considered a disadvantage to the new industrialized world. As working spaces required higher levels of light, architecture became a matter of an economic solution, where the demand of reducing the floor to floor height resulted in the usage of daylighting being insufficient. the light level requirements for industrial buildings grew to a point of 1000 lux at working stations, which was building the future for using electrical lighting only. (Phillips 2004)

“It is inevitable that artificial light must be- come the primary light source where efficiency of vision is combined with an economic analysis of building function. Natural lighting is becom- ing a luxury.” (Phillips 1964, p. 13)

the 1960’s proved to be the peak of the electrical lightings history when artificial lighting was not only the primary source of lighting but also became the dominant source. governmental energy policies de- cided that the electrical cost for lighting was far less than building with daylight if the requirements of light levels had to be accomplished. It was even up for debate that lighting should be the source for heat- ing for buildings. this approach was led by engineers and influenced many architects into agreement. Ar- chitects in the 1960’s, believed that the planning of light levels and how the nature of the electrical light could achieve these requirements were the first and most important decision to make, when designing a building, and so they began to disregard daylight as a functional source.

luckily, today we are open for new knowledge on how to use daylight as a functional light source, and use this inevitable natural phenomenon for our own advantage when it comes to architecture.

even though light from the sun arrives on the earth every day, daylight does not come for free when im- plemented into buildings. A window provides the opening into the interior, however there is a lot of different aspects associated with having a window; the control of the sunlight, the ventilation, the radiation from the sun, and the view which all affect the indoor

2.0 theoretiCal

baCkground

(10)

10

climate that is needed to be considered in order to make a pleasant indoor environment.

A new ‘genre’ of architecture today is called “passive architecture”, where buildings are structured to reduce the mechanical energy cost for light, heat and sound to a minimum. this new concept is also called the ‘net Zero energy demand’ and its main beliefs is that the energy consumed equals the energy harvest.

the centre of this new philosophy is sunlight and daylight which appears in greater or lesser amount every day and has a powerful affect for lighting in and on a building. First, daylight and sunlight provides the interior of a building with light through apertures, and second is the affect light has on the impact with the facade of a building that turns the light into energy as solar heat. this requires a highly technological window element that is able to restrict or harvest the exterior energy by selecting the windows’ coating, cavity fills, photovoltaics and advanced blind system and controls. All these aspects are in consideration when planning a building, which results in the need of seeing daylighting as a holistic approach to design. (Phillips 2004)

2.1.2 dAylIgHt VARIABIlIty

one very important aspect of daylighting is that it has infinite varieties of appearances. daylight has both in daylit indoor and outdoor spaces the ability to change. the human body is capable to adapt to different scenarios through our vision, while changes in daylighting is testing our adaptation technique on a daily basis. our perceptual attention is focusing, to a degree, towards changes:

• daylit changes that appear in an interior space as altering with time.

• changes in our daylit environment creates a continuous study of the space we are situ- ated.

• the obvious difference in the experience of a space lit entirely by artificial sources during daytime, where there is no access to daylight.

• the changes as clouds move through the sky, covering and revealing the sun as an inherent variability of daylight.

these are all part of our perceptual ritual that we meet every day and without the influence of the daylight we could in theory be without windows. Study have shown that the variability in daylight is affecting the eyes in a relaxing manner (Sze-Hui Au 1999). this natural way of renewal is part of the photochemical processes of the eye where it adapts to adjust for the changes in daylight.

the natural change from day to night provides a variety in light intensity, colours, contours and contrast towards darkness where the need for artificial sources to take over is necessary to orientate. that is the final stage in adaptation of the eye before the day starts again.

the weather has also an important role and the changes associated with changes from bright sunny days to dark and cloudy or rainy days are clearly im- printed in the indoor space. In the northern parts of the world it is common to all, that there is nothing as waking up to a rising sun in the morning on a bright

day, which is an experience that is lacking when the outside is dark and gloomy.

the changes in the weather is closely associated with the changes of the seasons. the winter brings snow, dark and cloudy weather where inhabitant of the north try to catch every last piece of the sunlight available. the summer brings bright and sunny weather and bring the shadows back to life that are casted in through the windows. As human beings, we perceive each season in our own way, but what is interesting is the way the outside world is read and experienced through the window, which provides the necessary information of the exterior world and al- low slight changes in the appearance of the indoor space. (Phillips 2004; Sze-Hui Au 1999)

2.1.3 dAylIgHt AVAIlABIlIty

the sun creates daily and seasonal movements with respect to a geographic location on the earth and because the sun is moving in the same orbit, it possible to predict the pattern of amount and di- rection of available daylight. A variation caused by changes in the weather, temperature, and air pollu- tion lies on top of this predictable pattern.

the many different types of daylight makes the calculations significantly more complex, compared to electrical lighting. to determine the illuminance level when daylight reach an aperture it is crucial to include the vary characteristics of the sky and sun, and the altering spatial connection between the sun and daylight apertures.

Internal illuminance levels is formed by two main factors:

- building geometry – the size of windows, the various internal reflectance, obstruction, etc.

- external daylight availability (dA) – defined by external illuminance levels, sky luminance distribution.

the annually and diurnally changes and as weather conditions alters, creates a constant variation in daylight levels. Additional to time variations is the geographical variation, which even change from region to region within the same country and has a significant value to daylight availability. geographical variations consist partly of latitude related variations that arise according to the solar altitude of the sun’s position in the sky to the horizon. A study made by chro- scicki (1971) shows that the average daylight data received (global horizontal) from warsaw, malmö, Stockholm, and kiruna result in the same curve of illuminance against solar altitude.

daylight availability is different from the daylight factor (dF) because the specific location, time, date and sky condition are all considered when calculating daylight availability. dA and dF are similar in that they both express the ratio of the indoor and outdoor illuminance level, but dA includes more details, which makes the outcome unique for a specific location.

Researchers have made measurements of daylight illuminance for the past 60 years on various locations all over the world, which have resulted in a very similar daylight availability data. these data and the equations derived from them, do not provide instantaneous values of illuminance and luminance

daylighting

(11)

11

but express mean values. these instantaneous values may differ greatly from those determined by calculated methods based on daylight availability, but where instantaneous data can be more than twice or less than half the mean design value, the equations give the best result from data received over time measurement sessions.

daylight availability data is crucial when designing a building with ideal daylighting solutions. calculated daylight availability for a space, provides data that contain:

(1) average illuminance conditions (hourly data for months or seasons)

(2) maximum and minimum illuminance conditions on vertical and horizontal surfaces

(3) clear and overcast sky design illuminance

values

(4) probability of a given illuminance value and the cumulative probability that a given value will be exceeded

(5) correlation of illuminances to irradiance values.

the daylight availability calculations begins with the determination of the solar position, which is a function of the latitude and longitude of a given space, Julian date (day of the year), and local time.

time is converted to solar time, angles are calculated to give the sun’s position in the sky. the specific sky condition is observed, and finally the daylight availability equation is used to set the daylight illuminance (IeSnA 2000; kensek & Suk 2011; Secker & littlefair 1987).

“Because it is difficult to judge the quantity of light, lighting design must be based on what one is able to perceive and what one wants to look at-the quality of the luminous environment.” (Lam 1986, p. 9)

Figure 3 | käthe kollwitz | mutter mit totem Sohn | neue wache | 1937-1938

(12)

daylighting

2.1.4 dAylIgHt HeAltH

the spectral characteristics of sunlight are affecting a variety of human physiological mechanisms either in a direct or indirect mean. the affected bod- ily functions are triggered by light transmitted from photoreceptors in the retina to neural signals. the reactions of skin to ultraviolet radiation is by far the most common and least significant of the different responds humans have to direct sunlight. overexpo- sure of sunlight will lead to tanning and eventually burning of the skin. However, the most beneficial advantage of exposure to sunlight is the skins production of vitamin d when absorbing ultraviolet radiation.

the human body’s absorption of calcium depends on the production of vitamin d, where lack of vitamen d can result in rickets in children and osteoporosis in adults. the advantageous of ultraviolet radiation has been researched over the years and shown how UV light can help the immune system and prevent certain diseases arising. low levels of vitamin d and absence of UV radiation leads to bone disease, cardiovascu- lar disease, infection disease, and cancer progno- sis. low exposure to UV radiation is also associated with a greater risk of autoimmune diseases (sclero- sis), th2-driven disease (asthma), non-obese diabetic (type 1 diabetes), and chronic inflammatory diseases (psoriasis) (Hart 2012).

In the northern hemisphere where winter brings along darkness and cloudy weather, people tend to use artificial illumination that simulates the solar spectrum. meteorological measurements shows that clouds are covering the sky for up to 83% of the year

(Jørgensen and cappelen 2015), which means the amount of ultraviolet radiation is lower and can only be compensated artificially through vitamin d added food or special UV light sources. the solar UV spectrum in tanning appliances, are equal to the amount of UV emission of the sun. Powerful solaria appliances are even up to 10 – 15 times higher than that of the midday sun (International Agency for Research on cancer (IARc) 2005) elderly or institutionalized adults who are getting insufficient sunlight would be able to meet their biological needs through artificial exposure of UV radiation up to an acceptable level in order to obtain the optimal amount of light (lam 1986).

one of the first documented principles of how architects should use daylight in buildings, is in Vitru- vius’ de Architectura (1st century Bc). He describes through ten books, classical architectural principles of harmony proportion and symmetrical methods.

Among all these code of conduct, Vitruvius empha- size the healthy aspects of daylight and how it affects the indoor environment in residential and farming houses. He focus on the location and latitude of the site according to the different kind of “human races”

living in the north and in the south. He describes the orientation of the rooms towards the movement of the sun and how the function of the room requires different exposures of light (See Appendix). these guidelines are fundamental for designing buildings with integrated daylighting, and important in order to create a healthy indoor space. even though we have moved a little further in the knowledge of daylight

Figure 4 | United cigar Stores company | Saving daylight | USA | 1908

(13)

13 daylighting

we now see the consequences of poor daylighting and absences of sunlight. ‘Seasonal Affected disor- der’, SAd or fall-winter depression is a syndrome that has its name from the lack of light, at the one time of the year when the days are short and the light is limited. SAd is characterized by recurring around fall and winter, where the people affected by SAd tend to feel typical depression symptoms as depressed mood, low energy, withdrawal from social activity and no initiative. In addition, the people affected by this syndrome experience atypical vegetative symptoms such as prolonged sleep (hypersomnia), and increased appetite (Rosenthal et al. 1984).

when working in an office, the connection to the outside world can be limited. According to a research, people have an innate longing to be in contact with nature (white and Heerwagen 1998). the need of being able to have a view to the outdoor is only possible through windows. An additional explanation for wanting a view to the outdoor may be that the view creates a source of environmental stimulation.

the use of open plan office structure is a way of increasing productivity by concentrating the work- flow, but the cost of planning so, is the possibility of minimum to zero of daylight on a workplace. the productivity level in an office with inadequate working conditions and poor lighting will deteriorate which lead to a work output that may decay compared to the gain of positive results (Phillips 2004).

Having a workplace close to a window is without doubt preferable and studies has proven that people tend to be more productive when experiencing positive emotional states, and by providing their preferred workplace their positive emotional state will increase.

(wright and cropanzano 2000).

lighting designed in a poorly manner can affect the health of the people at a working place and lead to stress and various forms of discomfort such as, dry or itching eyes, migraines, aches, pains and other disadvantage. these are all part of Sick Building Syndrome, where lighting plays an important role in people’s well-being. An investigation of the lighting in working environment have suggested that natural lighting in a work space was associated with fewer reports of headache and lethargy during work (Rob- ertson et al.1989).

working our way from how lighting affects the environment into how lighting is affecting our mental health, this next section will be concerned with how light is influencing the human circadian rhythm.

even with or without the respect for light’s ability to stimulate human vision it has always been used as a noun when it came to describe the stimulus, even more than describing the response from a biological system.

Figure 5. illustrates the suprachiasmatic nucleaus (Scn) in the hypothalamus of the human brain. Rea et al. 2010.

located in the suprachiasmatic nuclei (Scn) of the hypothalamus, (see figure 5), in the mammal brain are endogenously driven physiologic fluctua- tions generated by a master neural clock, also called the circadian rhythm. the retina in the eye reacts to a temporal light and dark pattern that is controlling the Scn and is for humans, to some extent, greater than 24 hours. the rotation of the earth around its axis creates a light and dark pattern that the retina synchro- nizes to the Scn to match a 24-hour day. disruption of the natural light and dark pattern, through rotating shift work and fast flying across time zones leading to a wide range of complaints and contribute to the development of illness such as; poor performance, in- somnia, increase in weight and risk of breast cancer., which have been confirmed by research.

Since research can prove that the light and dark pattern reaching the retina have a great influence on the human health and well-being, it is even more important to integrate both light and dark as stimuli to the human circadian rhythm. It has recently been tested and accepted that short-wavelength light, or blue light, affects the circadian rhythm with a peak sensitivity close to 460 nm, when testing the suppres- sion of melatonin or phase shifting of the dim light melatonin onset (dlmo). A hormone that is known as the stress hormone, cortisol, is part of the circadian rhythm as the body produce a high level of cortisol in the awakening period. this is also known as the cortisol awakening response (cAR). the short-wavelength light enhance the cAR in humans and stimulates the body to know, when it is time for it to be active, and could help adolescents preparing for any environmental stress they might experience (Rea et al. 2010;

Figueiro and Rea 2012).

(14)

14

daylight is remarkable as a light source because of its unique, variable spectra and distribution and its contribution to the dynamic spatial experience of the everyday environment. daylight is a powerful natural resource that has significant impact on human health and the ability to increase human satisfaction, and at the same time, it can sustain energy, but only when certain aspects as glare control, human factors, view design, and integration of building systems is treated in a proper manner. (Hopkinson et al. 1966; Illuminat- ing engineering Society of north America (IeSnA) 2000)

the light received from the sun appears in two types of lighting: daylight as direct sunlight, and daylight as indirect (ambient) light. the ultimate daylight source is the sunlight while the skylight as ambient light consists of diffused sunlight caused by the atmosphere, seen as the blue sky, and is the primary illumination in an interior space. A supplementary illumination from the reflected sunlight from the ground and surroundings is also affecting the daylighting of an interior space. Following is given three types of

light sources created by the sun and the sky (Hopkin- son et al. 1966; Illuminating engineering Society of north America (IeSnA) 2000):

the sun as a light source

due to the rotation of the earth around its axis, and its position to the sun, there is created an apparent motion of the sun connected to any point on the surface of the earth. the relation between any given point on the earth and the sun is described in terms of two angle: the solar altitude, which is the suns vertical angle to the horizon also called the “Solar Zenith An- gle”, and the solar horizontal angle from due south (in the northern hemisphere) called the

“Solar Azimuth Angle” (see Figure 7).

the sky as a light source

the sky luminance is the traversed sunlight, which has been scattered by dust, water vapour and other particles in the atmosphere and assisted by clouds. the sky is categorized into three qualities:

Figure 6 Sky luminance mapping with the three different cIe standard sky luminance distribution - 1. clear sky, 2. Partly cloudy sky (intermidiate), 3.

overcast sky. (kenny et al. 2006 printed online at www.pro-lite.co.uk)

2.2 sun & sky as sourCes of light

(15)

15

clear, partly cloudy, and overcast. when the sky is not overcast it can then be subdivided into three grades of covering of the sun, obscured, partly obscured, and clear.

the ground as a light source

the different ground surfaces can also have an influence on the daylighting design. the light that is received on the ceiling or walls, depending on the colour and material, reflected onto other interior surfaces. Around 10 - 15% of the total daylight fenestrating a window onto a daylight elevation is light reflected from the ground, and can increase when the ground consist of a light-coloured surface as sand or snow. when a window is shaded the light reaching the window will then mainly consist of ground reflected light.

Figure 7. the sun’s position in terms of solar altitude (a1) and azimuth (aS) with respect to the cardinal points of the compass. By inspiraton of IeSnA (2000).

2.2.1 Sky lUmInAnce

the luminance distribution of the sky is essential for making daylighting calculations. mapping out the different skies under a variety of occurrences from clouds covering the whole sky to cloudless and clear situations with or without the sun’s influence, is a way of standardizing the sky in order to predict different daylight pattern (kittler and darula 2002). Figure 6. shows sky maps from dublin of the three c.I.e. sky luminance distributions (kenny et al. 2006), figure 8 and figure 9 shows such mentioned sky luminance maps of Stockholm made by Hopkinson, R.g. (1954)

that daylight gives a peculiar satisfaction by its many perceptual varieties is a general thought but it creates complications when it comes to standardizing the sky luminance in the design of interior daylighting. the methods for measuring daylighting design are based on the same techniques as of those used for measuring artificial lighting design. many aspects of the design of the room affect the amount of light received in the room. the size, position and orientation of the window and the luminance of the sky are all aspects that need to be considered. different surfaces from interior decorations and wall, floor and ceiling are setting the amount of light that is uti- lized by inter reflection. we can change the window design and the interior of a room but we are not able

to control the sky luminance (Hopkinson et al. 1966;

lam 1986).

As described previous in this thesis under section 2.1.3 daylight Variability, the constant changing variations of daylight is caused by the apparent position of the sun, which is a connected with the time of the day, season of year, and the position of the building in terms of latitude, longitude, and orientation. other consequences of random variations as the gathering and passage of clouds of water vapour or dust, and of industrial haze and fog are resulting in calculation being made on a statistical basis. If we ignore the cIe standards of sky luminance for a moment, we can describe the daylight that varies with time, and because of the sun’s changing position on the sky and to different cloud cover we are not able to justify that the sky luminance is constant nor uniform. In order for making lighting calculation of daylight it is necessary to separate direct sunlight from skylight and approach the two conditions differently. the direct sunlight is recognized as a point source and the sky is seen as a large diffuse source with different luminance distribution (Hopkinson et al. 1966; lam 1986).

A method for capturing the necessary data to create a set of standard skies started in 1760 when lam- bert published Photometria, but it was not until the beginning of the 20th century, that several scientists started to take interest in sky luminance distribution and the idea of using an overcast sky as reference for lighting calculation was introduced in a research made by P. moon and d.e. Spencer in 1942, later tested by R.g. Hopkinson and finally adopted in 1955 by the c.I.e. (c.I.e. 1955) It was prior to the assumption that the overcast sky was creating uniform luminance and would therefore be the base for all daylight calculations of an overcast sky conditions and thus became standard that could be used in all parts of the world.

Figure 8 Sky luminance for

clear sky, 11 am, Stockholm Figure 9 Sky luminance for overcast sky, 11 am, Stockholm

many researches have been focused upon finding a newer and more precise solution for measuring and predicting the sky conditions, which brings us to the latest standard that the c.I.e. has adopted.

A method that describes fifteen different sky types of relative luminance distribution are based on daylight measured data at Sydney, Berkeley and tokyo.

the standard is researched by kittler et al. in 1998 and confirmed by c.I.e. in 2003 (c.I.e. 2003), where five overcast, five clear and five transitional skies are set to cover more detailed sky luminance distribution. the whole incidence spectrum where different diffuse scattering by the atmosphere are considered and including the possible effects of direct sunlight.

the standard model allows calculations of illuminance and luminance levels for both relative and ab-

sun & sky as sourCes of light

(16)

Figure 10 | Unknown artist | Sunbeams | the Sun magazine | 2014

(17)

17 sun & sky as sourCes of light

solute physical units because it is possible to apply additional parametrisation (kittler et al. 1998).

even with these new and more detailed sky distribution types, the computational simulation is still using the two basic sky luminance distributions as calculation methods for daylight design (kensek, k.

and Suk, J.y. 2011):

The Clear Sky - where the blue sky and the position of the sun are considered and the relevant atmosphere scattering is constant. the luminance of the ground and vertical surfaces from the illuminance of the sun and sky needs to be considered. It is also necessary to know the properties of all the surrounding reflecting surfaces.

The C.I.E Standard Overcast Sky - where the sky is completely covered with clouds and is considered as of uniform luminance distribution. the effect of the direct sun is excluded.

2.2.2 cleAR Sky

the air mass surrounding the earth consist of molecules and dust particles which are scattering and absorbing some of the sunlight traversing the atmosphere. the scattering of the sunlight is more commonly seen as shorter wavelengths, or as the blue colour on a clear sky. As a result of the light scattering, the dark blue seen at the edge of the atmosphere is the blackness of space, a deep blue at an altitude from the highest mountains, a light blue at sea level, and when the haze of water vapour and dust is encountered it is perceived as almost white. the sky has even a higher colour temperature than the direct light from the sun, which means that the sky is seen

“warm” or “blue”.

the direct light from the sun creates most of the illumination on a clear day but the intensity of the sunlight varies according to the thickness of the air mass it passes through. the intensity of light from the sun is at any latitude less at sunrise and sunset, but differ at noon because the sun’s position is lower at higher latitudes. Since the air mass varies little during a day, the distinction in the sun’s intensity from over- head to 15 degrees above the horizon is insufficient, at which time the intensity has decreased by 50% of maximum. 10% of the total clear-day illumination comes from the brightest area of the sky immediately surrounding the sun being sunlight “trapped” by the atmosphere and becomes a non-uniform luminous source. the darkest part of the blue sky is about 300 footlamberts, which is equal to clouds on an overcast day, (and is approximately 90 degrees from the sun.)

the clear sky distribution may not be preferable in parts of the world where clouds are covering the sky most of the year, e.g. in the Scandinavian countries.

the clear sky model would be more suitable in pre- dominant sunny climates, e.g. countries surrounding the earth’s equator. though, this model would be useful for studying visual glare and thermal discomfort created by the sun anywhere in the world (Hopkinson et al. 1966; lam 1986).

2.2.3 oVeRcASt Sky

the other most extreme sky condition is when the sky is completely covered with clouds also called overcast sky. this type of sky is the one, which reduce the sunlight by more than 90%, in which it is the criti- cal design condition of all the sky condition models.

For a sky can be categorized as an overcast sky the condition of the cloud cover need to be so dense that they create a full obscuration of the sun. In such a condition, the luminance level is the same, at any point of the same elevation above the horizon.

through measurements made on a fully overcast sky obtained in different parts of the world has shown the average luminance of the sky is brightest at the zenith and systematically decreases to one-third of the zenith value at the horizon. Such a certain sky with a symmetrical luminance distribution about the zenith has come to be known as the “c.I.e. Standard overcast Sky”

the average illuminations from an overcast sky depend on the density of cloud cover and the altitude of the sun and can vary over a wide range. there has not been any measured illumination above 2.000 lm/

ft2 ~ 21.527 lm/m2 from an overcast sky and for the condition of a fully overcast sky to be fulfilled, the sun needs to be sufficiently obscured from dense mist, cloud, or dust, which result in an illumination which generally becomes very low. It is mentionable that while the average level of illumination as described before is depending on the altitude of the sun, the distribution of the sky luminance is symmetrical about the zenith and therefore not depended of the height of the sun. (Hopkinson et al. 1966; lam 1986).

2.2.4 tHe dAylIgHt FActoR

It has now been demonstrated how the many aspects of daylight is providing physiological and psychological environments surrounding our lives here on this planet. with knowledge from clarification of the different types of sky conditions, it is possible to move forward and define the daylight illuminated interior of a room. the general reason for this is to calculate the amount of daylight in an interior room, derived from a hemisphere of uniform luminance even though this would never happen under the changing circumstances of a real sky. the c.I.e. defined in 1955 the relative luminance distribution of a standard overcast sky, because of the extensive measurements of the sky distribution made from different measurement in various countries. this has led to the measurement value of the daylight factor.

the definition of the daylight factor is a method that describes the ratio of a daylight illuminated point on a given plane in an interior space illuminated directly or indirectly from a sky of assumed or known luminance distribution and the illumination of a point on a horizontal plane from an unobscured hemisphere of this sky. the direct light from the sun is excluded. the daylight Factor is partly a measure of the total daylight illumination of a point in an interior space of a building that receives direct daylight from the sky, and partly a total of any reflected light that may reach into the room through the window from an external surface, e.g. an opposite building, the ground, etc. and the light which is reflected and

(18)

18

Sky Component (SC) - Direct illumination from the sky

Externally Reflected Component (ERC) - External reflected illumination from the sky

Internally Reflected Component (IRC) - Internal reflected illumination from the ground, ceiling and walls Internally Reflected Component (IRC) - Internal reflected illumination from the floor, and ceiling

inter reflected from the interior surfaces in the space reaching the same reference point. Referring to the illustration in figure 11.

the daylight factor method is commonly used with uniform or c.I.e. overcast skies and mostly in the northern parts of the world where overcast skies are dominating the sky most of the year. the daylight Factor can be separated into three components that describes the different kinds of daylight which reach a point on a horizontal plane within an interior space.

the sky component (Sc) is the relation between the daylight illumination at a reference point on a given plane which is affected directly from a sky of an expected or known luminance distribution and the illumination on a horizontal plan which is illuminated from an unobstructed hemisphere of this sky. the sky component is used when measuring the direct daylight for both the c.I.e. clear Sky and the c.I.e overcast Sky but is not suitable for an uniform sky and unglazed openings. the direct sunlight is excluded for both sky conditions.

the externally Reflected component (eRc) is the relation between the daylight illumination at a reference point on a given plane which is affected from external reflecting surfaces illuminated directly or indirectly by a sky of an ex-

pected or known luminance distribution and the illumination on a horizontal plan which is illuminated from an unobstructed hemisphere of this sky. Any sunlight affecting the luminance of both the external reflecting surfaces and the comparable plane are to be excluded.

the Internally Reflected component (IRc) is the relation between the daylight illumination at a reference point on a given plane, which is affected from internal reflecting surfaces of the sky of an expected or known luminance distribution and the illumination on a horizontal plane which is illuminated from an unobstructed hemisphere of this sky. Any sunlight affecting the luminance of both the internal reflecting surfaces and the comparable plane are to be excluded.

the total daylight illumination at a reference point in an interior space is the sum of these three components. whereas daylight specialist used to calculate all different component adding them up to become the total illumination from daylight, equal to the daylight factor, the modern advanced technology has simplified the determination of the total daylight illumination to be calculated using computation simula- tions (Hopkinson et al. 1966).

sun & sky as sourCes of light

Figure 11 the different ways daylight reaches a point in a room. By inspiration of Hopkinson et al. 1966 p. 69

(19)

(20)

Figure 13 | william Smith | Art 3 | 2014

(21)

21

the word perception describes many areas of the human brain and we have difficulty explaining what perception truly is, but one thing is certain, that there is no such thing as objective perception.

this chapter will focus on the physical stimuli that is the ‘food’ for the perception ‘engine’ in our brain.

the activities of the sensory organs, which are not connected to the visual system, will not be mentioned even though they are also a part of the human perception. Perception is the receiving process that happens

‘behind the curtains’ in our brain and perception does not just happen by accident. the idea of the world surrounding us is the same world we perceive, is an in- nocent belief. this is probably a very harsh accusa- tion, but nevertheless true, which will be explained throughout this chapter.

the question of how we see the world around us is theoretical, factual and practical at once and the theories needed to explain seeing are within the history of psychology and philosophy. the facts needed to explain the theories come from psychology, phys- ics, and physiology, and the applications needed to test seeing extend to art, aviation, and photography.

the world we perceive is built on fundamental vari- ables that are spatial and temporal, where the world as we know it extends and endures (gibson 1950a)

the question of perceptual adaptation, which could have occurred through evolution, maturation, learning, alternatively short-term or momentary ad- justment is not being considered in this thesis.

the fundamental understanding of perception is how perception as a biological function, controls our actions and adaptive behaviour, developing in an evolutionary process through time.

the theories about human perception and action, has usually been investigated as logically independent, where a general description of a scene including objective quantities such as size, distance, shape, colour, and motion has become the goal of perception to prove a basis for any subsequent behaviour. For perception where vision is in the context of action, theoretically significant implication starts to appear (warren 1995).

Visual patterns in the surrounding environment of human observation creates visual stimulation that can be turned into reliable information of the three- dimensional layout. this phenomenon contributes to confusion because the properties of optical stimulation seem to have almost nothing in common with the real objects in a physical space. compared to the real objects, which is constructed of physical substances,

perceptual awareness of objects comes from stimulation, which is the flickering pattern of light, reaching the back wall of the eye. these patterns of light are the only information that many animals, including humans, need to create a layout of surfaces in a three- dimensional space. the relation between our perceptual knowledge of an environment and its physical structure has been an intriguing question central to the perceptual research for centuries. two theories have long been the main argument concerning our visual perception. one group of investigators believe that our visual perception derives from truthfulness while others discuss visual perception to be inherently insolvent and subject to large distortions. those who argue for the view that perception is being accurate points out how we successfully orientate us around objects in a world in motion and how we can perform any other visually guided behaviour. the group who argue that perception is impoverished, focus on how perceptual illusions has proven otherwise. what makes these arguments even more complicate is that both theories contain undeniable evidence. Percep- tual judgement has through well-controlled experiments proven distortion and accuracy in our visual perception.

An explanation for these two theories can be based on the discoveries of a german mathematician named Felix klein. klein proposed a categorisation of geometrical transformations of objects, where some will alter part of its structural properties and others will be invariant. this accounts also for the geomet- ric transformation between physical and perceived space. Perceived geometrical objects in the visual field, which are invariant, should be of higher accuracy. Perceived geometrical objects which are not invariant should result in systematic errors.

examining the relationship between physical and perceived space not only focuses on the information described above, but also all the potential sources of information provided by the natural vision. However the specification of the different aspects of object structure in the visual field does not need to be identi- cal.

It is without a doubt that the general source of optical information for perceiving three-dimensional structure is caused by motion and stereovision. these aspects have been extensively investigated during the past several years (todd et al. 1995), but it is therefore currently more interesting to examine the monocular vision and how it affects our visual perception.

2.3 perCeption

“The question is not what you look at, but what you see.”

(Thoreau, H.D. 1906. p. 373)

(22)

22

Vision helps almost all animals interact around in their environments. Special perceptual features are required for animals to find their way around, search- ing for food, seeking shelter, avoiding predators and many other activities. these perceptual features are to be found in the spatial layout of the visible environmental surfaces, as sizes, distances, shapes, and orientations and this is referred to as visual space perception. A perceptual feature such as judging the size of an object may be useful in several occasions, which has lead this feature to be thought of as tak- ing place independently for the particular action at the moment. one way of looking at space perception is that a chair may have a particular perceived size in a living room that is more or less independent of whether a person is sitting on it or standing on top of it.

this perspective describes “space perception” as existing independently of the specific activity of the person in the space – another perspective describe the situation more accurately as the person doesn’t perceive the size of the chair but instead experiences what activities is possible with that chair i.e. “I can sit on the chair” or that “I can stand on the chair”. It would possibly be more correct to assume that what is perceived is the behaviour given by what the environment affords, also called an affordance (gibson 1977; gibson 1979, p. 18), as an alternative for the physical appearances of the environment (warren 1995, p. 264). that some “higher” animals, as the pri- mates, are in some way combining the two theories is possible, which makes them perceive their environment connecting both the underlying layout and the affordance of this same layout. Since neither of the perspectives are validly compelling arguments, the both combined, result in a larger variety of information about space perception.

the features of a spatial layout that an animal perceives seems relevant for its behaviour even if the perception of the spatial layout is to some extend independent of the activity of the animal. this can be described as - for the animal to interact with its environment, the animal’s perception adapts to the environment. though, it is not possible to understand the space perception of the animal from its behaviour and the environment to which it is adapted. while it seems as the adaptability of our species makes our potential behaviours and environments unlimited, it seems too feasible to apply this theory to the space perception of humans (Sedgwick 2010).

the theory that light is reflected from the environmental surface into the retina of an observer and perceived as a three-dimensional spatial layout can cause a problem when it comes to visual space perception. then it is no longer only the behaviour and environment of an observer that is the main cues for

the space perception. A method for controlling the reflected light by its interaction with the environment has been made by J. J. gibson, where he calls the reflected light for an optic array. different optic ar- rays are produced by different environments which connects a specific structured optic array to the specific environment that produced it. this theory suggest that the optic array is carrying visual information from that related environment which makes it to some degree possible to work backwards and recover the structure of an environment from its related optic array – a process called inverse projection (gibson 1966).

there can only be one spatial layout, which has produced a particular optic array for inverse projection to be possible. consider that every spatial layout through any particular optic array requires only spatial layouts that adapt to the natural environment of the observer. It is necessary to identify the ecological constraints and to ensure that some structure in the optic array is determining certain visual information and confirming the validity of this information (cantril 1960, p. 41 cited in Sedgwick 2005; marr 1982, p. 104; Sedgwick 1983, p. 427).

2.4.1 tHe textURed gRoUnd

living mostly of our human lives on the surface of the earth has contributed towards humans developing as a terrestrial organism, which would cause the most basic spatial layout in what humans perceive as standing on a ground plane and looking towards the horizon. the optic array, in this spatial layout, consists already of visual information. the distance between locations on the ground and the observer are increasingly large. thus the optic array is optically projected to increasingly high angular elevations. By using a simple trigonometric relation, it is possible to link the distance along the ground to the angular elevation in the optic array, and to the height of the observer’s eye above the ground. An observer could quite accurately perceive the distance along the ground by using this angular height in the visual field (gibson 1950; epstein 1966; wallach and o’leary 1982). It has to be noticed that since the height of the observer is included in the relation with the angular elevation in the optic array, the distance increases with the eye height of the observer as the eye height is a natural unit of measurement, in compare to the observer’s own body.

defining the surface of the ground or floor as with a texture of some sort, creates a texture scale when observed, which can be used to perceive distance (gibson 1950). An observer moving around two objects resting on the ground will make the angular separation between the two objects vary with the po-

2.4 visual spaCe perCeption

(23)

Figure 14 | Vasilly kandinsky | orange | Bauhaus | 1923

(24)

24 visual spaCe perCeption

sition of the observer, but the amount of texture, or number of texture elements between the two objects will not change. distance between any two objects can be measured through the information the texture scale provides, also called exocentric distance, while the distance between an observer and an object is called egocentric distance. An example of an ecological constraint could be that the texture elements must have a statistically uniform distribution throughout the surface in order for the texture scale information to be valid. Several different articles concerning estimations of egocentric distance, specifies that results using behavioural methods, by observers choosing a location and with closed eyes walk to it, tend to be more accurate than psychophysically methods, or by obtaining verbal estimates of distance (gillam 1995; Sedgwick 1986; weist and Bell 1985).

this conclusion supports the previous hypothesis, that space perception is more familiar to affordances than to the reportable physical characteristics of the environment.

A method to measure exocentric distance perception done by researchers was performed by scattering a number of objects on a ground plane and asking a group of observers to draw down a map of the objects’ position on the plane or to guess the distance between the possible pairs of objects. the common tendency of the observers from this experiment was that the perceived distance from the observer and along a radial line was compressed by 15 to 50% relative to a distance estimated in the frontal plane of the observer (levin and Haber 1993; toye 1986; wagner 1985).

even small distances are possible to perceive without a ground plane, but the larger the distance is between the object and the observer, the more difficult the distance of objects is to perceive accurately if the location of the ground or some equivalent plane is not possible for the observer to locate. An example would be to estimate the distance of an unknown object in the sky, in the absence of having the ground as a reference plane, the object could be large and at a far distance or be much smaller and be much closer.

the importance of a reference plane as a continuous ground for perceiving distance has been demonstrated through research, where an experiment showed that the distance perceived was less accurate when there was a gap in the ground between an object and an observer (Sinai et al. 1998).

the scale of the ground texture and the position of the horizon is what specifies the size of an object standing still on the ground. the far distance to the horizon makes the line of sight nearly parallel to the ground and makes this line intersect all objects, which is equal to the eye height of the observer. It is possible to position the horizon in the optic array even if it is not visible. other optic array information, such as vestibular information, or linear perspective can specify the position of the horizon.

Another parameter of affordances is the horizon- ratio relation¹, which express how the eye calculates the size of a doorway to pass through or the height of 1 The horizon-ratio relation (sedgwick 1973 cited in sedgwick 2005; sedgwick 1983): The height (“s”) of the object, is relative to the eye height (“h”) of the observer, is approximately equal to the optic array angle (“s”) subtended by the object, relative to the optic array angle (“h”) subtended by the portion of the object below the horizon: s/h = s/h (figure 15).

a platform to step up on. this type of affordances are scaled by an observer’s eye height to the particular size of the body. observers’ ability of using this information for planning or performing actions is very precise and has been proven through research (Ji- ang and mark 1994; warren 1984; warren and whang 1987).

Figure 15 the horizon-ratio relation, by Sedgwick 2005

It is possible to misperceive the size of an object and their affordance if the observer is relying only on the horizon-ratio relation and the observer is placed on a box which increases his eye height (mark 1987;

wraga 1999), but the error cancels out if several objects are visible (Sedgwick 1983). Research has proposed that when an object is in the same height as the observer’s eyes, the size perception based on this information is most accurate (Bertamini et al. 1998).

2.4.2 occlUSIon

objects are often scattered around a typical environment and to help specify the spatial relation between these objects are a wide range of visual information. From one perspective of observation are surfaces which are seen either fully visible; partially covered or fully covered by other surfaces; or fac- ing the away from the observer. that one surface is partially covering another, creates information about a relative distance of the two surfaces from the observer. this is also called a partial occlusion and the surface being partially occluded will necessarily be further away. even though partially occlusion informs the observer that there is a distance between the two objects, it does not provide any information of the certain size in depth between the two objects, but it does let the observer knows that the surface being occluded must be at least the thickness of the oc- cluding surface farther away. this is also known as the order of depth.

For a surface to be partially occluded the observer needs to know that there is more of the surface that is visible. Identifying or familiarizing some objects or forms is important for seeing the covered part of a surface. But how is it possible, that we still are able to understand unfamiliar forms and objects even though they are partially covered? one major aspect of occlusion comes into account in the meeting of two objects; the passing of a surface behind another, result in a break in the contours of the occluded surface where the two contours meet (Helmholtz (1962/1925); Ratoosh 1949) this obstruction of contours in the optic array is called a t-junction because of the continuing contour forms the crossbar of the letter t, and the covered contour forms the stem of the letter t (guzman 1969). An example of such, when two contours create a t-junction, is shown in figure 16. while figure 16 shows a classical example of occlusion of two contours, other contour character-

(25)

25 visual spaCe perCeption

istics, such as figure 17, can also contribute to the perceptual elements of occlusion, even with three- dimensional volumes.

2.4.3 context

An environment of surfaces creates the context for other surfaces, and may be perceived relative to this context in terms as size, shape, and location of that particular surface (Sedgwick 1986). to give an example as shown in figure 18, a filled circle on the side of a large cube may be perceived as smaller than a filled circle of equal size on a smaller cube, because the first filled circle is smaller relative to its context. In very complex environments the context is important in the perception of spatial layout. even so, context is a complex matter and has not yet been thorough- ly investigated. when it comes to how texture scale and horizon-ratio provides information for size, it has been discussed whether they are specific instances of contextual influences. the contact between two surfaces are a necessity for creating contextual information for some forms. An example would be that, perceiving the size of two smaller surfaces depends on the texture scale of the background and the contact of any edge of the surfaces with the background (gibson 1950a, p. 181; gillam 1981).

Another form of visual information is cast shadows, which can either confirm or disconfirm, physical contact (madison et al. 2001). A simple example of cast shadow is when an object is lifted above the ground, which result in a visible gab between the bot-

tom of the object and the shadow on the ground. this gap between the edge of an object and its own cast shadow provides the observer with the information of the object not being in physical contact with the ground and founding the spatial relation between the suspended object and the ground (Rock et al. 1982;

Ujike and Saida 1998; yonas et al. 1978).

when the optical projection of an edge is overlay- ing on top of the optical projection of another surface, it is perceived as the edge is in physical contact with another surface, which is called optical contact (gibson 1950a, p. 178). even if a surface is raised in space and there is no physical contact between the two surfaces it may still be optical contact. An example where the optical contact would occur without physical contact is highly unlikely but possible.

within the same line of sight it is possible that two projected contours meet but is in reality separated in space, figure 19a. the example is highly uncom- mon since the eye of sight change when shifting even slightly around the object in focus, which will reveal a visible gap between the endpoints, figure 19b. thus it is only a small part of all the points of observation which could produce optical contact. A general position assumption (Huffman 1971), is a another example of an ecological constraint used for describing when there is no qualitative change in the optic array when changing the position of one’s viewpoint. this assumption helps through wide application the understanding of perception (nakayama and Shimojo 1992; Rock 1983). Visual illusions are violating the general position and are typical created on a static

Figure 16 t-junction for occlusion of contours Figure 17 the perception when one object penetrate another

Figure 18 the context of what we perceive changes our judgement

of sizes Figure 19 optical contact and context. (a) Perception of contact. (b)

no physical contact (a)

(b)

(26)

Figure 20 | James turrell | Aten Reign | guggeheim new york | 2013