Content uploaded by Hyoyoung Ahn
Author content
All content in this area was uploaded by Hyoyoung Ahn on Nov 04, 2022
Content may be subject to copyright.
Odd and Even in Lighting Design:
Aesthetic Research Using Symmetry
Hyoyoung Ahn
Light in Performance
Rose Bruford College
This dissertation is submitted for the degree of
Master of Arts
October 2022
Acknowledgements
I would like to say thank you to my family for supporting me and all participants of the
survey across various industries.
ii
Abstract
This paper studies the advantages of composing visual elements with odd numbers
rather than even numbers through comparing odd cases and even cases in lighting
design. It also surveys whether people prefer odd or even numbers of arranged lights in
stage lighting. Much evidence suggests that using odd numbers in visual composition has
been preferred for a long time in visual design, including photography, film, graphic design
and painting. This preference is evident in each artistic area; however, the disciplinary and
theoretical evidence is not sufficiently strong to support it. The theoretical basis for the
aesthetic value of odd numbers is formed by combining the research about symmetry and
aesthetic preference from various scientific areas with an analysis of the trend in visual
design. The effects of lights arranged in odd and even numbers are compared via
experiments. This research explores whether the universal aesthetic preference is
applicable in visual design and provides a new guideline for lighting design.
Keywords: neuroaesthetics, odd and even, visual attention, lighting design,
symmetry, stage lighting
iii
Table of Contents
1. Introduction ..................................................................................................1
2. Literature review ..........................................................................................3
2.1 How to evaluate objects .............................................................................3
2.2 Preference for symmetry in faces and the environment .............................4
2.3 Preference for odd numbers ......................................................................7
3. Methodology ................................................................................................9
4. Experiments and results ............................................................................10
Experiment 1: Comparison between lights in rows .........................................11
Question 1-1: Which one do you prefer or looks better? ................................11
Question 1-2. Why did you think so? .............................................................11
Question 1-3. Where did your gaze mainly fall on each picture? ...................12
Experiment 2: Comparison between lights in chess board arrangements ......13
Question 2-1. Which one do you prefer or looks better? ................................14
Question 2-2. Why did you think so? .............................................................14
Question 2-3. Where did your gaze mainly fall on each picture? ...................15
Experiment 3: Comparison between lights on proscenium sets .....................16
Question 3-1. Which one do you prefer or looks better? ................................17
Question 3-2. Why did you think so? .............................................................17
Question 3-3. Where did your gaze mainly fall on each picture? ...................18
Question 4. How much are you conscious of odd and even when you watch
visuals? .........................................................................................................19
Question 5. How much are you conscious of odd and even numbers when you
design visually ...............................................................................................19
Statistics for the responses 'neither was preferred' ........................................20
5. Discussion .................................................................................................21
6. Conclusions ...............................................................................................25
Appendix 1: Table of Data Summary ..............................................................28
References ....................................................................................................29
iv
1. Introduction
In many visual design areas, the viewers’ long-term attention is important. If
someone observes a certain object continuously, this should be because it arouses a good
feeling. Designers in all fields related to visuals have been making various efforts to attract
customers to their productions. Using odd numbers is one of the common designing
methodologies passed by word of mouth. The tendency to prefer odd numbers is
discussed in many websites, books and articles in industries such as photography, film,
graphic design and painting (Bourne, 2022). In arts, the methodologies used to create a
piece of art are mostly based on experiences; this is not often derived from disciplinary
research.
However, since neuroaesthetics, the new scientific study that uses neuroscience to
explain humans' aesthetic experiences, was introduced by Zeki (1999), the previously
unknown reasons for the public's inclination to the trend have gradually been revealed.
This inspired me to research why odd numbers are preferred in visual design. Biology,
psychology and geometry and their theoretical backgrounds provide much evidence to
explain this phenomenon. This research hypothesises that odd numbers look better,
especially in lighting design. This is suitable for experimentation because lights have a
large visual impact depending on their numbers in lighting design.
Traditionally, lighting design has been taught focusing on functional perspectives
and practice in the field. Lighting design methodologies are not often disciplinarily
explained like in other artistic areas. The lighting design books published in the market are
usually written technically and functionally. Due to this, it is taught with a technical rather
than artistic approach in Asian countries, including South Korea. The trend seems
emphasised because many lighting designers and lighting directors in South Korea have
studied electronic engineering before working on lighting design. Surely, technology is
important in lighting design. For instance, film is a type of art that changes as new
1
technologies are developed, and learning more about the operation and function of the
equipment is considered important (Malkiewicz and Mullen, 2005). Also, lighting design
has extended to a type of art that can create new images beyond just the ability to light an
incandescent light as Thomas Edison did or create a movable light as Vari Lite did (Brox,
2010; Moody and Dexter, 2017, p. 19).
The purpose of this study is to present a new way of lighting design through
disciplinary support. In previous studies, some research can be found about the
relationship between beauty, symmetry and visual attention. However, direct academic
research about odd and even numbers is few in visual design, particularly in lighting
design. Ultimately, a new theory is produced to apply to actual lighting practices by lighting
designers in various fields, including performance, architecture and film. Thus, this study
will prove meaningful by providing a new perspective on aesthetic criteria in arts.
2
2. Literature review
This literature review covers papers in design, psychology, neuroaesthetics and
biology to find scientific bases for the hypothesis that odd numbers can better capture the
public’s attention than even numbers. Also, this section aims to apply the theoretical
background to lighting design as a visual design through reviewing the fields, including the
rest of lighting design, because there are few studies in lighting design related to this.
2.1 How to evaluate objects
People evaluate objects using specific criteria commonly explained as visual
hierarchy, assessing by scale, colour, contrast and placement (Lupton and Phillips, 2015,
p. 299). This is done to judge whether the object is something we like or dislike, beneficial
or harmful (Houwer, 2009). For instance, when a person finds a shirt larger than their body
while shopping, they assess the shirt to determine whether they prefer it over a well-fitting
one. If they prefer Coca-Cola over Pepsi, someone will pay attention to red drinks in a
supermarket. When choosing seats in a restaurant, someone prefers a brighter seat over
a darker one.
In the same context, people tend to subconsciously find human faces when looking
at objects because our brains have evolved to quickly sense others’ presences and their
identity or emotional status (Adams et al., 2011, cited in Palmer and Clifford, 2020). The
Gestalt theory proposes that if objects are placed closely, they will be recognised as one
cluster; this may help explain the phenomenon of recognising a human face when an
object possesses visual elements that look like two eyes and a mouth placed in the same
position as on a human's face (Treisman, 1982; Ben-Av et al., 1992). As viewers use
personal criteria to assess the values of an object, their preferences will affect how they
feel about the arrangement and shape of visual elements in visual design. Just as people
3
subconsciously prefer certain objects, they may also subconsciously prefer odd or even
numbers in visual design.
2.2 Preference for symmetry in faces and the environment
Researchers determined several criteria that people use to assess whether a
human face is attractive or not. The most preferred face is symmetrical in terms of the
distance between the eyes and the length of the eyebrows based on the nose’s centre.
This was supported by the experiment conducted by Chatterjee (2011) with infants and
phenomena in wildlife cited by Ryan (2018). This phenomenon is visible across cultures
and ages. It implies that the preference for symmetrical faces is not only related to cultural
experiences (Chatterjee, 2011, cited in Ryan, 2018, p. 75). People prefer people likely to
pass on healthy genes to future generations, with strong resistance to pathogens, and
facial symmetry indicates this characteristic. As the extent of facial symmetry increases, so
does genetic quality, including growth rate, fecundity and survivability (Little et al., 2011,
cited in Thornhill and Gangestad, 1994). In the same context, Thornhill and Gangestad
(1993) insisted that averageness is an important element in distinguishing whether a face
is attractive or not. Because average faces have more various genes than less-average
ones, they have genetic benefits, such as proteins that pathogens like parasites cannot
adapt to (Thornhill and Gangestad, 1993). The preference for superior genes significantly
contributes to evaluating facial attractiveness.
4
Fig 1: Comparison between asymmetrical faces and symmetrical faces. (a, c) an asymmetrical proportion in eyes, nose,
eye-brows and (b, d) a perfect symmetrical proportion (Modified from Generated Photos, 2022).
(a)
(b)
(c)
(d)
This phenomenon can be found easily in the environment. Bilateral symmetric
shapes are found in most human-friendly animals, including dogs, crabs and elephants,
because symmetry is very important from a health perspective (Moubayidin and
Østergaard, 2015). For example, having symmetry in physical elements like arms and legs
makes great balance and is advantageous to movement. If some parts of the body are
bigger, with imbalanced proportions, it could indicate a problem for the tissues supporting
the body; for example, bones may break more easily due to the imbalance of the body.
Thus, it could be said that symmetry has been naturally preferred because it is a stable
shape that can maintain physical health.
Safety is essential for basic human living; people avoid dangerous animals and eat
non-poisonous plants to satisfy their appetite. This danger avoidance is a natural response
that has been learned. Because of this, people naturally feel physiological repulsion in
response to objects they judge as dangerous. Trypophobia is the best example of this; it is
typically formed by an aversion towards clusters of small holes. The triggers of
trypophobia include bubble wrap, lotus seed pods, insect eyes and condensation, and it
results in fear and anxiety, itching and emotional distress (Alberts, 2021). Trypophobia is
caused by very dangerous animals, including crocodiles, poisonous snakes and spiders,
having skins with repeating circular patterns. Our ancestors were afraid of such patterns
because such animals pose a threat, and they passed on these characteristics to their
5
Fig 2: Silhouettes of animals having symmetry and
asymmetry. (a) a crab with asymmetrical proportion
on each hand and (b) shows same length of a dog's
legs (a: Modified from Macrovector, 2022, b: Modified
from tsign703, 2022).
(a)
(b)
descendants, so the disgust continues today (Le et al., 2015). Creatures having many
interactions with humans also show this tendency. This demonstrates that the shapes that
dangerous creatures have are not preferred. Thus, it could be said that a symmetrical,
stable shape is instinctively preferred. If so, symmetry can be considered a very generally
preferred shape. This raises the question of what the most important visual elements to
consider are when identifying the extent of symmetry between two halves of an object.
The elements distinguishing symmetry are the nose in the face and the midrib in a
leaf. Because the nose is located in the middle of the face and the midrib is placed in the
centre of the leaf, it becomes a reference point that enables us to identify symmetry
between the left and right sides. It can be used not only on faces and leaves but also when
looking at almost all objects. Gazing at the centre area first when looking at an object
helps distinguish the most accurate shape of it; in this way, symmetry can also be
efficiently grasped (Desanghere and Marotta, 2015). Consequently, people tend to identify
symmetrical shapes when looking at faces and nature because they prefer symmetry.
The element in the exact middle of an object plays an important role in distinguishing
symmetry. In a situation where the visual elements are as one group parallelly arranged
with odd numbers, the odd configuration has the visual element in the centre always (Fig.
4). Odd numbers always have a visual element which can be focused on the middle,
6
Fig 3: A leaf's symmetrical
proportion (Modified from
Macrovector, 2022).
Fig 4: Comparison in odd & even
arrangements (Ahn, 2022a).
enabling us to easily grasp their symmetry. An odd-numbered design can capture more
attention than an even-numbered one with no visual elements at the centre. This supports
the hypothesis that odd numbers look better in a lighting design.
2.3 Preference for odd numbers
Much research about mathematical odd and even numbers can be found. For
example, when subjects were asked what their most preferred number was, about 30%
chose the number seven (Zubair and Sunny, 2015). However, there are few disciplinary
materials about odd and even numbers related to visuals. Thus, research from many
areas about preferences for odds and evens was reviewed.
The multi-award-winning painter Adams (2015) highlighted that using odd numbers
is beneficial in painting because it makes viewers focus on the centre of the picture. The
tendency to arrange and group elements by odd numbers in Japanese rock gardens was
revealed through Tonder and Lyons’s (2005) study. The ‘Rule of Odds’ is a famous
composition in cinematography, configuring actors or objects in a frame to show as odd
numbers by combining them into three groups; Bourne (2022) found that this can give a
better aesthetic impression. In the food industry, it has been hypothesised that there is a
tendency to prefer food plated in odd numbers; however, researchers’ findings regarding
7
Fig 5: The Danaïdes (Waterhouse,
1906).
Fig 6. Odd numbered food plating
(Begel, 2022).
this were inconclusive (Hernandez, 2015; Woods et al., 2016). Unlike other areas that
enable us to assess just visually, the research struggled to compare the preference for odd
or even objectively because when comparing the same amount of odd and even food by
adjusting the size of it (same weight but they served three pieces and four pieces), one of
the two would be reported as more preferable because of more amount of food can affect
to the subjects’ appetites.
The preference for odd numbers is presented across various fields of visual
directing. However, there are few mentions of the preference for even numbers,
disciplinarily or in any industry. Many cases, like the food plating one, hypothesise that odd
is better; however, not much research about the trend has not been carried out. This study
aims to fill this research gap.
8
3. Methodology
The experiment was processed using an online Google form survey comparing the
lighting design pictures generated by Capture, the stage lighting visualisation programme,
that asked which designs were preferred. The survey was distributed by email to Rose
Bruford College students and posted on the researcher's Facebook and Instagram
profiles. The survey required participants to specify their preferred design from three sets
of lighting pictures showing odd and even numbers of lights. It then asked participants why
they preferred each selection and where their gaze usually fell on each picture. The survey
also asked participants if they were conscious of the odd and even numbers when they
consumed visual media, if they worked in a visual-related area (e.g. lighting, set, theatre
blocking , graphic design, photography or film) and how much they were conscious of odd
1
and even numbers when they designed.
The collected data were analysed to determine whether they supported the
hypothesis. The reasons for participants’ preferences were collected via essay questions
and classified into several similar categories. They were sorted to ‘another aesthetic
reason’ or ‘etc.’ if they were difficult to categorise. I analysed the commonalities and
differences between preferences for odd and even numbers by determining where the
participants mainly gazed when looking at each picture. The data from the participants
who had no preference was separately examined by extracting from the three cases
because the numbers were very small. The difference between the answers of visual
workers and ordinary individuals was analysed. The preferences for odd and even
numbers were summarised by individually analysing each question.
Blocking: Positioning actors on a stage in a performance, including theatre, musical, opera
1
9
4. Experiments and results
This experiment was conducted using a qualitative research-based survey. The
survey's title was ‘The preference about odd number and even number in lighting
composition’ to avoid bias about the hypothesis. I explained the purpose of this research to
the participants in the survey. I asked them why their preferred choice looked better, why
they did not prefer the other option, and how much they were conscious of odd and even
numbers when they designed visuals, including lighting, set, theatre blocking, graphic
design, photography and film, to get meaningful results.
Participants
This survey was conducted from 7 September to 14 September 2022. A total of 105
individuals (58 males, 43 females and four individuals who did not report whether they
were male or female) participated in this experiment. Their occupations included stage
lighting experts (31% or 33 individuals) and entertainment industry workers (32% or 34
individuals). The rest of the participants (36% or 38 individuals) held various other
occupations. Fifty-seven per cent of the participants were South Korean, 12% were British,
10% were American, and 20% were from various other countries (mostly European).
Stimuli
Four and five lights were used in even and odd cases, respectively. Four and five
are the smallest numbers at which the odd and even compositions are distinguishable.
Using four and five minimised the influence caused by a difference in amount, as it
becomes difficult to identify odds and evens if the number is over six.
10
Experiment 1: Comparison between lights in rows
Participants were presented with a picture of five lights (odd) in a row and a picture
of four lights (even) in a row and asked to choose between the two.
Question 1-1: Which one do you prefer or looks better?
In Question 1-1, 62 individuals (59%) chose the odd case, 39 individuals (37%)
picked the even case, and four responded that neither was preferred. The percentage
favouring the odd case is 1.59 times higher than the even case. This arrangement was the
simplest configuration that allowed participants to identify whether the case was odd or
even. It was the most preferred case among the three experiments.
Question 1-2. Why did you think so?
11
Fig 7. Comparison between 5 lights in a row and 4 lights in a row for Experiment 1 (Ahn, 2022b).
59%
37%
(62)
(39)
Odd selectors
37%
6%
10%
18%
29%
Allowing gaze at centre
Stability
Balance
Symmetry
Another aesthetic reason
Even selectors
36%
3%
15%
18%
28%
Fig 8. Two pie charts for the reasons about Question 1-2 in odd selectors and even selectors.
(23)
(18)
(6)
(11)
(14)
(1)
(7)
(11)
(6)
(4)
Question 1-2 asked why the participants chose their answers to Question 1-1. This
was designed to determine whether their reasons for their preferences were consistent
with the literature. Eighteen individuals who chose the odd case reportedly thought it was
better because it allowed them to gaze at the centre of the image. Eleven individuals
attributed their choice to stability, six to balance and four to symmetry. Twenty-three
individuals attributed their choice to another aesthetic reason. In the odd case, ‘it allowed
me to gaze at the centre’ was the majority answer (29%) except for ‘another aesthetic
reason’, followed by ‘stability’ (18%), ‘balance’ (10%) and ‘symmetry’ (6%). In the even
case, 11 Individuals reported it was better because it was more balanced. Seven attributed
their choice to symmetry, six to stability and just one reported that it allowed them to gaze
at the centre. Fourteen individuals said their choice was due to another aesthetic reason.
In the even case, ‘balance’ was the majority answer (28%) except for ‘another aesthetic
reason’, followed by ‘symmetry’ (18%), ‘stability’ (15%) and ‘it allowed me to gaze at the
centre’ (3%). In summary, those who preferred the odd case tended to do so because it
allowed them to gaze at the centre, and those who chose the even case mostly did so
because of the image’s balance.
Question 1-3. Where did your gaze mainly fall on each picture?
12
Even selectors
10%
3%
5%
8%
10%
21%
44%
Odd selectors
2%
3%
5%
8%
13%
69%
Centre light or centre area
The overall image
Beam areas
Between lights
Spot areas
etc
Each edge light or area
Fig 9. Two pie charts for gazing points about Question 1-3 in odd selectors and even selectors.
(2)
(1)
(5)
(8)
(43)
(8)
(17)
(3)
(1)
(4)
(4)
(3)
(2)
Question 1-3 asked about the participants’ gazing point when looking at each
picture in Question 1-1. This was asked to determine the difference between the gazing
points of odd and even selectors. Forty-three individuals choosing the odd case reported
they mainly gazed at the centre light or centre area. Eight individuals responded with ‘the
overall image’, five with ‘beam areas’, three with ‘between lights’, two with ‘spot areas’
2 3
and one with ‘each edge light or area’. In the odd case, ‘centre light or centre area’ was the
majority answer (69%), followed by ‘the overall image’ (13%), ‘beam areas’ (8%), ‘between
lights’ (5%), ‘spot areas’ (3%) and ‘each edge light or area’ (2%). In even selectors, 17
individuals mainly gazed at the centre light or centre area. Eight individuals responded
with ‘beam areas’, four with ‘each edge light or area’, three with ‘the overall image’, two
with ‘between lights’ and one with ‘spot areas’. Four individuals answered in various other
ways. In the even case, ‘centre light or centre area’ was the majority answer (44%),
followed by ‘beam areas’ (21%), ‘each edge light or area’ (10%), ‘beam areas’ (8%),
‘between lights’ (6%), ‘spot areas’ (2%) and ‘etc.’ (10%). In summary, odd and even
selectors both tended to gaze at the centre light or centre area.
Experiment 2: Comparison between lights in chess board
arrangements
I performed the second experiment by showing each participant two pictures. The
first depicted a light arrangement consisting of five lights horizontally and five lights
vertically (5x5). The second was a picture of four lights arranged horizontally, with four
lights arranged vertically (4x4). I asked participants to pick the arrangement they preferred.
The odd case’s arrangement was five by five even though it did not look similar to the
even case to keep the consistency in this experiment.
Beam: An appearance of a path of light, between a light source and a spot point.
2
Spot: An area lit by a light.
3
13
Question 2-1. Which one do you prefer or looks better?
In Question 2-1, 51 individuals (49%) chose the odd case, 49 individuals (47%)
picked the even case, and five individuals responded that neither was preferred. Two more
individuals selected the odd case than the even case, but the preference percentages
were almost equal.
Question 2-2. Why did you think so?
Question 2-2 asked why participants chose their selection in Question 2-1 to
determine whether their reasons were consistent with the literature, even in a complex
format. Thirteen individuals choosing the odd case reported it was better because it
allowed them to gaze at the centre. Eleven individuals said it ‘looks rich’, eight chose ‘feels
spatial’, one of each chose ‘symmetry’ and ‘balance’, and ‘stability’ respectively. Sixteen
14
Fig 10. Comparison between 5x5 square arrangement and 4x4 square arrangement for Experiment 2 (Ahn, 2022c).
49%
47%
(51)
(49)
Odd selectors
31%
2%
2%
2%
16%
22%
25%
Allowing gaze at centre
Looks rich
Feels Spatial
Symmetry
Balance
Stability
Another aesthetic reason
Less complex
Even selectors
25%
2%
6%
8%
21%
38%
(16)
(13)
(11)
(8)
(1)
(1)
(1)
(1)
(4)
(18)
(3)
(10)
(12)
Fig 11. Two pie charts for the reasons about question 2-2 in odd selectors and even selectors.
individuals answered that their reason was due to another aesthetic reason. In the odd
case, ‘it allowed me to gaze at the centre’ was the majority answer (25%) except for
‘another aesthetic reason’, followed by ‘looks rich’ (22%), ‘feels spatial’ (16%), and
‘symmetry’, ‘balance’ and ‘stability’ (2%). In even selectors, 18 individuals reported that the
even case was better because it looked rich. Ten individuals responded with ‘stability’, four
with ‘symmetry’, three with ‘balance’ and one with ‘it allowed me to gaze at the centre’.
Twelve individuals answered that their reason was due to another aesthetic reason. In the
even case, ‘looks rich’ (38%) was the majority answer except for ‘another aesthetic
reason’, followed by ‘stability’ (21%), ‘symmetry’ (8%), ‘balance’ (6%) and ‘it allowed me to
gaze at the centre’ (2%). In summary, odd selectors tended to pick ‘it allowed me to gaze
at the centre’ and ‘looks rich’; even selectors tended to select ‘less complex’
Question 2-3. Where did your gaze mainly fall on each picture?
Question 2-3 asked the participants where their gazing point was when looking at
each picture in Question 2-1 to determine the difference between gazing points in odd and
even selectors. Thirty-one individuals choosing the odd case reported they mainly gazed
at the centre light or centre area. Six individuals responded with ‘between lights’, five with
‘beam areas’, four with ‘spot areas’, two with ‘the overall image’, the other two with ‘each
15
Odd selectors
2%
4%
4%
8%
10%
12%
61%
Centre light or centre area
Between lights
Beam areas
Spot areas
The overall image
Each edge light or area
etc
Even selectors
4%
4%
4%
6%
18%
20%
43%
Fig 12. Two pie charts for gazing points about question 2-3 in odd selectors and even selectors.
(1)
(2)
(2)
(4)
(5)
(6)
(31)
(2)
(2)
(2)
(3)
(9)
(10)
(21)
edge light or area’ and one with ‘etc.’. In the odd case, ‘centre light or centre area’ was the
majority answer (61%), followed by ‘between lights’ (12%), ‘beam areas’ (10%), ‘spot
areas’ (8%), ‘the overall image’ (4%) and ‘each edge light or area’ (4%). In even selectors,
21 individuals mainly gazed at the centre light or centre area. Ten individuals responded
with ‘the overall image’, nine with ‘between lights’, three with ‘spot areas’, two with ‘beam
areas’, the other two with ‘each edge light or area’ and the other two with ‘etc.’. In the even
case, ‘centre light or centre area’ was the majority answer (43%), followed by ‘the overall
image’ (20%), ‘between lights’ (18%), ‘spot areas’ (6%), ‘beam areas’ (4%), ‘each edge
light or area’ (4%) and ‘etc.’ (4%). In summary, odd and even selectors tended to gaze at
the centre light or centre area.
Experiment 3: Comparison between lights on proscenium sets
This experiment was designed in a way that is used in a general proscenium
theatre to provide an example similar to designs found in actual theatres. The odd picture
4
showed five lights at the top and two columns of five lights on each side. The even picture
showed four lights at the top and two columns of four lights on each side. The pictures
were shown together, and participants were asked to choose between the two. Testing
with the format used in general proscenium theatres is the most valuable example
because it applies to actual lighting design in theatre.
Proscenium theatre: A traditional theatre form having a frame in front of the stage.
4
16
Fig 13. Comparison between a odd-numbered proscenium set and a even-numbered proscenium set for Experiment 3
(Ahn, 2022d).
57%
31%
(60)
(33)
Question 3-1. Which one do you prefer or looks better?
In Question 3-1, 60 individuals (57%) chose the odd case, 33 individuals (31%)
picked the even case, and 12 responded that neither was preferred. The percentage
favouring the odd case was 1.81 times higher than the even case's percentage.
Question 3-2. Why did you think so?
Question 3-2 asked participants why they made their selection in Question 3-1 to
determine whether their reasons were consistent with the literature when viewing a
realistic stage lighting example. Twenty individuals choosing the odd case reported it was
better because it ‘looks rich’. Eleven individuals responded with ‘feels spatial’, eight with ‘it
allowed me to gaze at the centre’, three with ‘stability’ and one with ‘balance’. Seventeen
individuals answered that the reason was due to another aesthetic reason. In the odd
case, ‘looks rich’ was the majority answer (33%), followed by ‘feels spatial’ (18%), ‘it
allowed me to gaze at the centre’ (13%), ‘stability’ (5%) and ‘balance’ (2%). In even
selectors, 12 individuals reported that the even case was better because it was less
complex. Four individuals responded with ‘balance’, three with ‘feels spatial’, two with
‘symmetry’, and the other two with ‘stability’. Nine individuals answered that the reason
17
Odd selectors
28%
2%
5%
13%
18%
33%
Looks rich
Feels spatial
Allowing gaze at centre
Stability
Balance
Less complex
Symmetry
Another aesthetic reason
Even selectors
27%
6%
9%
9%
12%
36%
Fig 14. Two pie charts for the reasons about question 3-2 in odd selectors and even selectors.
(17)
(1)
(3)
(8)
(11)
(20)
(12)
(9)
(3)
(2)
(4)
(3)
was due to another aesthetic reason. In the even case, ‘less complex’ was the majority
answer (36%), followed by ‘balance’ (12%), ‘feels spatial’ (9%), ‘symmetry’ (9%) and
‘stability’ (6%). In summary, odd selectors tended to pick ‘looks rich’, while even selectors
tended to select ‘less complex’.
Question 3-3. Where did your gaze mainly fall on each picture?
Question 3-3 asked participants where their gazing point was when looking at each
picture in Question 3-1 to determine the difference between gazing points in odd and even
selectors. Twenty-eight individuals choosing the odd case reported they mainly gazed at
the centre light or centre area. Ten individuals responded with ‘the overall image’, six with
‘beam areas’, the other six with ‘spot areas’, four with ‘each edge light or area’, one with
‘between lights’ and five with ‘etc.’. In the odd case, ‘centre light or centre area’ was the
majority answer (47%), followed by ‘the overall image’ (17%), ‘beam areas’ (10%), ‘spot
areas’ (10%), ‘each edge light or area’ (7%) and ‘between lights’ (2%). In even selectors,
13 individuals mainly gazed at the centre light or centre area. Six individuals responded
with ‘the overall image', the other six with ‘each edge light or area’, four with ‘spot areas’,
one with ‘between lights’ and five with ‘etc.’. In the even case, ‘centre light or centre area’
was the majority answer (39%), followed by ‘the overall image’ (18%), ‘each edge light or
18
Odd selectors
8%
2%
7%
10%
10%
17%
47%
Centre light or centre area
The overall image
Beam areas
Spot areas
Each edge light or area
Between lights
etc
Even selectors
3%
9%
12%
18%
18%
39%
Fig 15. Two pie charts for gazing points about question 3-3 in odd selectors and even selectors.
(1)
(5)
(4)
(6)
(6)
(10)
(3)
(1)
(6)
(4)
(13)
(6)
(28)
area’ (18%), ‘spot areas’ (12%), ‘between lights’ (9%) and ‘beam areas’ (3%). In summary,
odd and even selectors tended to gaze at the centre light or centre area.
Question 4. How much are you conscious of odd and even when you
watch visuals?
Twenty-two individuals responded that they were very conscious, 33 picked
‘conscious’, and 55 answered ‘not conscious’. ‘Very conscious’ and ‘conscious’ responses
made up a combined 48% of the total, and ‘not conscious’ constituted 52%.
Question 5. How much are you conscious of odd and even numbers
when you design visually
This question was given to visual workers. Thirty-seven individuals responded that
they were very conscious, two picked ‘conscious’, and 25 picked ‘not conscious’. ‘Very
conscious’ and ‘conscious’ responses made up a combined 61% of the total, and ‘not
conscious’ constituted 39%. The percentage favouring ‘very conscious’ and ‘conscious’
was 1.56 times higher than ‘not conscious’.
19
52%
27%
21%
Fig 16. A pie chart of question 4.
(55)
(22)
(28)
39%
3%
58%
Very conscious
Conscious
Not conscious
Fig 17. A pie chart of question 5.
(25)
(2)
(37)
Statistics for the responses 'neither was preferred'
These are combined statistics about the response ‘neither was preferred’ in
Questions 1-2, 2-2 and 3-2. Fig. 18 shows why the participants had no preference for odd
or even arrangements. When asked why they had no preference, eleven individuals
responded with ‘no difference’, eight with ‘another aesthetic reason’ and two with ‘etc.’. ‘No
difference’ was the majority answer (55%), followed by ‘another aesthetic reason’ (35%)
and ‘etc.’ (10%). Fig. 19 illustrates the gazing points in Questions 1-3, 2-3 and 3-3. Ten
individuals mainly gazed at the centre light or centre area. Four individuals responded with
‘beam areas’, two with ‘spot areas’, one with ‘the overall image’, one with ‘each edge light
or area’, one with ‘between lights’ and two with ‘etc.’. ‘Centre light or centre area’ was the
majority answer (48%), followed by ‘beam areas’ (19%), ‘spot areas’ (10%), ‘the overall
image’ (5%), ‘each edge light or area’ (5%) and ‘between lights’ (5%).
20
10%
35%
55%
No difference
Another aesthetic reason
etc
Fig 18. A pie chart for the
reasons(question 1-2, 2-2, 3-2) of all
responses from nothing were
preferred.
(11)
(8)
(2)
10%
5%
5%
5%
10%
19%
48%
Centre light or centre area
Beam areas
Spot areas
The overall image
Each edge light or area
Between lights
etc
Fig 19. A pie chart for the gazing
points(question 1-3, 2-3, 3-3) of all
responses from nothing were preferred.
(2)
(1)
(1)
(2)
(10)
(1)
(4)
5. Discussion
The collected data shows that an odd number is preferred when comparing lighting
arrangements with similar conditions but odd and even numbers of lights. Regardless of
preferences, almost all participants reported they mainly gazed at the centre light or centre
area in all experiments. This data indicates that odd numbers can be more attractive in
visual compositions because an odd-numbered arrangement has a visual element at the
centre to help the viewer distinguish symmetry.
The proportion of participants' occupations helped to obtain meaningful results
because it captured the perspectives of professionals and the ordinary public. In all
experiments, the participants responded that odd cases were overwhelmingly preferred,
except in Experiment 2. When asked about their preference, most odd selectors said they
preferred it because it allowed them to gaze at the centre. In addition, when looking at
each case, the response ‘gazing at the centre light or centre area’ was the most common
in all questions and experiments for odd and even selectors. Odd selectors gave it more
than even selectors. This finding aligns with the disciplinary principles previously
discussed; people tend to gaze at the centre of an object to find symmetry. This is
consistent with the hypothesis established in the literature review. However, the preference
for each case was split evenly in Question 2-1, with half of the respondents preferring the
even case and half preferring the odd one. Participants’ reasons for choosing were
contrasting. This result seems to be due to a difference in taste; some reported preferring
a ‘more rich’ arrangement and others a ‘less complex’ one. In other words, it could be said
that if an odd case were not complex, participants who normally preferred the even case
would instead select the odd one. I could change the number of lights from five to three to
experiment with similar conditions. However, this might reduce the experiment’s reliability
by disturbing the consistency of the entire experiment, introducing a new variable in
Experiment 2. It is likely that this split occurred because the numbers of the odd case were
21
maintained at five in Experiment 2 even though this made the odd and even cases look
dissimilar. Similarly, the odd case is much more preferred in Experiment 3 which was
conducted in the general proscenium theatre format. Experiment 3 was designed by
arranging odd and even numbers of lights on each top and side area. The odd case was
overwhelmingly preferred in this experiment, even though it was more difficult to
distinguish odd from even in the layout than in Experiment 1 because it had three sections.
While many ways of designing attractive lighting have been explored, academic
approaches have not been tried aggressively because the field is practice-based. The
existing related studies or materials could not clearly explain why visual workers prefer
odd numbers in their designs. After recognising symmetry as a universal aesthetic
preference inspired by neuroaesthetics, building a more logical understanding of the
preference for odd numbers was possible.
Light sources, beams or lit surfaces (spots) can be direct visual sources in lighting
design. As each dot is a visual source in pointillism , lights can be composed into visual
5
sources of various shapes such as dots, lines and surfaces when used in performance,
and lighting design has the characteristic of having a geometric shape like other visual
design fields. Therefore, some criteria used in many visual design fields, including stability,
balance and symmetry, are applicable. If each group consists of an odd number of
elements, they will likely look beautiful because that type of design shows stability. It does
not really have to be symmetrical; if it has an element that enables the viewer to gaze at
the centre of each group, it will be enough to capture the public’s attention. In lighting
design, because it is common for cases to finish as geometric shapes, using odd numbers
is one way to convey universal stability. Therefore, composing in odd numbers has a clear
advantage. In an interview with Robertson (1984), Jennifer Tipton said that most people
cannot be conscious about lighting, but all people are affected by it. In other words, the
Pointillism: a style of painting developed in France at the end of the 19th century in which a painting is
5
created out of small spots of pure colour that seem to mix when seen from far away.
22
public cannot easily be aware of the intention of the visual outcome. However, that means
eventually they are hugely affected by how visuals are designed, it can be one of the
important factors enabling re-consume. The participants in this study tended not to be
conscious of odd and even numbers. It can be inferred that the public also subconsciously
prefers odd numbers in composition. If practitioners follow this in their lighting design
working process, they can look forward to effectively capturing the audience’s attention.
Several limitations are observed in the experiment. First, the results were not
derived from objective judgment from the participants in the same environment; they
reported what they recognised. Therefore, it could be said that there is some lack of
accuracy about whether the gazing points they identified correlate with where their gazes
actually fell. More objective outcomes would be obtained if professional measuring
equipment, including an eye tracker, accompanied a future study. Second, the number of
given cases is small. Only parallel, chess board and proscenium arrangements were
presented in this experiment. Future research should attempt to raise the reliability of
hypothesis verification by including more cases, such as circular and multiple-line
arrangements. Finally, many more individuals must participate to increase the reliability of
the data.
Many artists are trying to break the existing frames, bringing unexpected pleasure.
However, the existing aesthetic theories can easily capture the public’s attention; they can
get expected pleasure from stability. Expected pleasure in visuals is a crystal of visual
patterns caused by positive emotions from what humans have seen. Because this has
been created through the experiences of our ancestors for thousands of years, it will
continue to make people feel universally impacted (Le et al., 2015). Usually, there is
conventional wisdom in society that the reason for aesthetic preferences cannot be
defined. It is surprisingly not difficult to find several clues that can already be explained by
evolutionary biology. If we can use inter-disciplinarily study to determine why the public
23
prefers some designs and why classical products are being survived, the people who
experience these stable designs will enjoy greater satisfaction in their lives, improving their
quality of life.
24
6. Conclusions
In this paper, I studied whether the preference for odd numbers, which has long
been a trend in visual works, including painting, film and food plating, is found in lighting
design. There were some clues related to this in neuroaesthetics and biology: visually
preferred shapes vary depending on the individual's background; however, symmetrical
forms were found to be visually preferred shapes in all cultures. This is commonly
observed when looking at faces. Symmetry can be seen as a universal criterion of
aesthetics because many animals and plants have symmetrical shapes. It is easy to
identify whether an object is symmetrical or not when gazing at its centre because that
allows us to grasp its shape. For instance, when arranging an even number of cubes with
at equal intervals in a row, there is no visual element at the centre. In contrast, when an
odd number of cubes are arranged at equal intervals in a row, it can easily attract the
public’s attention because the visual element in the centre makes it easier for viewers to
look at.
The experiment demonstrated my hypothesis based on the theoretical background
built in the literature review through several questions, including comparing odd- and even-
numbered lighting photos, why participants preferred one over the other, and where they
gazed in each photo. The collected data were compared based on whether participants
preferred the odd or even case, and the commonalities and differences were analysed.
Surprisingly, the results supported the hypothesis. It was found that participants preferred
odd cases because they allowed them to gaze at the centre. The same responses were
given in the general proscenium theatre case.
In summary, it can be said that the odd number is one of the cases enabling to
create more attractive shapes rather than saying the premise causing much visual
attention is odd number, even if it is not certainly arranged in odd numbers, because it can
evoke a sense of stability when the centre of an object or the centre of a group has visual
25
element. In addition, it is important to use recognisable numbers, whether odd or not, such
as three, five or seven, to meaningfully capture attention.
I suggested this as a method applicable to the lighting design working process. The
findings support that using odd in a group in visual composition enables the designer to
effectively hold the public’s attention based on the universal human preference for
symmetry. Since the invention of the proscenium stage, many performances have
empowered the centre of the stage. The main character is usually placed at the centre of
the stage or the centre of the crowd in a performance. It is a non-verbal expression
conveying that the character is the most important person. For example, kings and
presidents address from the centre of the stage, and the speakers of the House of
Commons in the UK conduct debates at the centre of the Commons chamber. Due to the
characteristics of theatres, stage lights can be hung on horizontal and vertical structures in
a way that clearly distinguishes whether an odd or even number of lights has been used.
Therefore, the audience can perceive the number of lights when lighting sources or beams
are directly exposed or surfaces are lit. Thus, if a lighting designer configures odd-
numbered lights based on the centre, the audience can better concentrate on the
performance because they feel a sense of stability visually. This effectively gives them
aesthetic satisfaction. Maintaining odds in lighting design is not always possible because
of limited budgets or issues with physical instruments. However, it can be applied
regardless of budget because at least a crucial area at the stage can have an odd number
of elements.
It was not possible to produce sufficiently objective results compared with surveys
conducted in a limited environment because this experiment was conducted by online
survey. Also, because it was conducted with only three lighting design cases, it will be
necessary to increase accuracy by comparing more cases in future studies. The
participants reported that they chose odd or even cases, respectively, because the case
26
looked richer or less complex. It could be found that the amount of visual stimulation is
also related to one of the significant reasons for preferring certain visuals. Inspired by this,
I feel it is necessary to study why the public prefers minimalism and maximalism in lighting
design based on some conditions, including cultural backgrounds and individual
personalities. If it is possible to summarise the conditions under which the preferences for
minimalism and maximalism are formed, it is expected that designers and productions will
be able to appeal more effectively to consumers.
27
Appendix 1: Table of Data Summary
28
Title
Question
Preference
1-1
Odd
59%(62)
Even
37%(39)
2-1
Odd
49%(51)
Even
47%(49)
3-1
Odd
57%(60)
Even
31%(33)
Reason
Odd selectors
1-2
Allowing gaze
at centre
29%(18)
Stability
18%(11)
Balance
10%(6)
Symmetry
6%(4)
Another
aesthetic reason
37%(23)
2-2
Allowing gaze
at centre
25%(13)
Looks rich
22%(11)
Feels spatial
16%(8)
Symmetry
2%(1)
Balance
2%(1)
Stability
2%(1)
Another
aesthetic reason
31%(16)
3-2
Looks rich
33%(20)
Feels spatial
18%(11)
Allowing gaze at
centre
13%(8)
Stability
5%(3)
Balance
2%(1)
Another
aesthetic reason
28%(17)
Even selectors
1-2
Balance
28%(11)
Symmetry
18%(7)
Stability
15%(6)
Allowing gaze at
centre
3%(1)
Another
aesthetic reason
36%(14)
2-2
Less complex
38%(18)
Stability
21%(10)
Symmetry
8%(4)
Balance
6%(3)
Allowing gaze at
centre
2%(1)
etc
25%(12)
3-2
Less complex
36%(12)
Balance
12%(4)
Feels spatial
9%(3)
Symmetry
9%(3)
Stability
6%(2)
Another
aesthetic reason
27%(9)
Gazing point
Odd selectors
1-3
Centre light or
centre area
69%(43)
The overall
image
13%(8)
Beam areas
8%(5)
Between lights
5%(3)
Spot areas
3%(2)
Each edge light
or area
2%(1)
2-3
Centre light or
centre area
61%(31)
Between lights
12%(6)
Beam areas
10%(5)
Spot areas
8%(4)
The overall
image
4%(2)
Each edge light
or area
4%(2)
etc
2%(1)
3-3
Centre light or
centre area
47%(28)
The overall
image
17%(10)
Beam areas
10%(6)
Spot areas
10%(6)
Each edge light
or area
7%(4)
Between lights
2%(1)
etc
8%(5)
Even selectors
1-3
Centre light or
centre area
44%(17)
Beam areas
21%(8)
Each edge light
or area
10%(4)
The overall
image
8%(3)
Between lights
5%(2)
Spot areas
3%(1)
etc
10%(4)
2-3
Centre light or
centre area
43%(21)
The overall
image
20%(10)
Between lights
18%(9)
Spot areas
6%(3)
Beam areas
4%(2)
Beam areas
4%(2)
etc
4%(2)
3-3
Centre light or
centre area
39%(13)
The overall
image
18%(6)
Each edge light
or area
18%(6)
Spot areas
12%(4)
Between lights
9%(3)
Beam areas
3%(1)
Consciousness
in usual
4
Not conscious
52%(55)
Conscious
27%(28)
Very conscious
21%(22)
Consciousness
in design
5
Very conscious
58%(37)
Not conscious
39%(25)
Conscious
3%(2)
Reason
Nothing selectors
1-2,
2-2, 3-2
No difference
55%(11)
Another
aesthetic reason
35%(8)
etc
10%(2)
Gazing point
Nothing selectors
1-3,
2-3, 3-3
Centre light or
centre area
48%(10)
Beam areas
19%(4)
Spot areas
10%(2)
The overall
image
5%(1)
Each edge light
or area
5%(1)
Between lights
5%(1)
etc
10%(2)
References
Adams, S. (2015) Three is the magic number. Available at: https://
www.sandyaskeyadams.com/blog/90252/three-is-the-magic-number (Accessed 21
August 2022).
Ahn, H. (2022a) Comparison in Odd & Even Arrangements Created by Photoscape X,
illus.
Ahn, H. (2022b, c, d) Generated Lighting Photos Created by Capture 2022, illus.
Alberts, N. (2021) What Is Trypophobia? Symptoms, causes, diagnosis, treatment, and
prevention. Available at: https://www.everydayhealth.com/trypophobia-101-
beginners-guide/ (Accessed 16 August 2022).
Begel, A (2022) Odd Numbered Food Plating, illus. Available at: https://
www.lifehacker.com.au/2022/03/16-of-the-trickiest-ways-restaurants-get-you-to-
spend-more-and-how-to-avoid-them/ (Accessed 10 September 2022).
Ben-Av, B. M., Sagi, D. and Braun, J. (1992) ‘Visual Attention and Perceptual Grouping’,
Perception & Psychophysics, 52, 3, 277–294. doi: 10.3758/BF03209145
Bourne, W. (2022) The rule of odds and what it means for your film. Available at: https://
www.videomaker.com/how-to/shooting/composition/the-rule-of-odds-and-what-it-
means-for-your-film/ (Accessed 9 September 2022).
Brox, J. (2010) Brilliant: The Evolution of Artificial Light, New York: Houghton Mifflin
Company.
Cambridge Dictionary (n.d.) Pointillism. Available at: https://dictionary.cambridge.org/
dictionary/english/pointillism (Accessed 3 September 2022).
Ryan, M. (2018) A Taste for the Beautiful: The Evolution of Attraction, Princeton: Princeton
University Press.
29
Desanghere, L. and Marotta, J. J. (2015) ‘The Influence of Object Shape and Center of
Mass on Grasp and Gaze’, Frontiers in Psychology, 6, 1537, 10. doi: 10.3389/
fpsyg.2015.01537
Generated Photos (n.d.) A.I Generated Virtual Faces, illus. Available at: https://
www.generated.photos (Accessed 8 September 2022).
Hernandez, M. (2015) Plate food like a pro. Available at: https://food-
hacks.wonderhowto.com/how-to/plate-food-like-pro-0161437/ (Accessed 9
September 2022).
Houwer, D. J. (2009) ‘How Do People Evaluate Objects? A Brief Review’, Social and
Personality Psychology Compass, 3, 1, 36–48. doi: 10.1111/
j.1751-9004.2008.00162.x
Le, T. D. A., Cole, G. G. and Wilkins, J. A. (2015) ‘Assessment of Trypophobia and an
Analysis of Its Visual Precipitation’, The Quarterly Journal of Experimental
Psychology. Volume 68, Issue 11. doi: 10.1080/17470218.2015.1013970
Little, C. A. and Jones, C. B. and DeBruine, M. L. (2011) Facial attractiveness:
evolutionary based research Review. Philosophical Transactions of The Royal
Society 366, 1638-1659. doi: 10.1098/rstb.2010.0404
Lupton, E. and Phillips, C. J. (2015) Graphic Design: The New Basics Second Edition,
Revised and Expanded, New York: Princeton Architectural Press.
Macrovector (n.d.) Autumn Leaves Set, illus. Available at: https://www.freepik.com
(Accessed 9 September 2022).
Macrovector (n.d.) Seafood Icons Set, illus. Available at: https://www.freepik.com
(Accessed 9 September 2022).
Malkiewicz, K. and Mullen, D. (2005) Cinematography: A Guide for Film Makers and Film
Teachers (3rd ed.), New York: Simon & Schuster.
30
Moody, L. J. and Dexter, P. (2017) Concert Lighting: Techniques, Art and Business (4th
ed.), New York: Routledge.
Moubayidin, L. and Østergaard, L. (2015) ‘Symmetry Matters’, New Phytologist, 207,
985–990. doi: 10.1111/nph.13526
Palmer, J. C. and Clifford, W. G. C. (2020) ‘Face Pareidolia Recruits Mechanisms for
Detecting Human Social Attention’, Psychological Science 2020, 31, 8, 1001–1012.
doi: 10.1177/0956797620924814
Robertson, N. (1984) Jennifer Tipton, the theater's magician of light. Available at: https://
www.nytimes.com/1984/02/11/arts/jennifer-tipton-the-theater-s-magician-of-
light.html (Accessed 25 August 2022).
Thornhill, R. and Gangestad, S. W. (1993) ‘Human Facial Beauty: Averageness,
Symmetry, and Parasite Resistance’, Human Nature, 4, 237–269. doi: 10.1007/
BF02692201
Thornhill, R. and Gangestad, S. W. (1994) ‘Human Fluctuating Asymmetry and Sexual
Behaviour’, Psychological Science, 5, 297–302. doi: 10.1111/
j.1467-9280.1994.tb00629.x
Tonder, G. and Lyons, M. (2005) ‘Visual Perception in Japanese Rock Garden Design’,
Axiomathes, 15, 353, 371. doi: 10.1007/s10516-004-5448-8
Treisman, A. (1982) ‘Perceptual Grouping and Attention in Visual Search for Features and
for Objects’, Journal of Experimental Psychology: Human Perception and
Performance, 8, 2, 194–214. doi: 10.1037/0096-1523.8.2.194
tsign703 (n.d.) German Shepherd and Dalmatian Dog Silhouette, illus. Available at:
https://www.freepik.com/premium-vector/german-shepherd-dalmatian-dog-
silhouette_3370635.htm (Accessed 9 September 2022).
Waterhouse, W. J. (1906) The Danaïdes, illus. Available at: https://artuk.org/discover/
artworks/the-danaides-108090 (Accessed 10 September 2022).
31
Woods, T. A. and Michel, C. and Spence, C. (2016) Odd versus even: a scientific study
of the ‘rules’ of plating. doi: 10.7717/peerj.1526
Zeki, S. (1999) ‘Art and the Brain’, Journal of Consciousness Studies, 6, 6–7, 76–96.
Available at: https://www.researchgate.net/publication/
233603316_Art_and_the_Brain (Accessed 23 September 2022).
Zubair, H. and Sunny, M. (2015) ‘Understanding Parity; Is the odd-effect odd or even?’,
European Conference on Visual Perception. August 2015. Available at: https://
www.researchgate.net/publication/281372044_Understanding_parity_Is_the_odd-
effect_odd_or_even (Accessed 16 September 2022).
32
