Available via license: CC BY
Content may be subject to copyright.
Original Research
Received: September 3, 2021. Accepted: October 6, 2021.
*Corresponding author: Marcos Roberto dos Reis. E- mail: mreis@uffs.edu.br; Eugenio Andrés Díaz Merino. E-mail:
eugenio.merino@ufsc.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
Product, Management & Development, 19(1), e20210010, 2021 | https://doi.org/10.4322/pmd.2021.005 1/12
Image composition influences on the mood board
visual reading process through eye-tracking
Marcos Roberto dos Reisa* , Eugenio Andrés Díaz Merinoa*
aPrograma de Pós-graduação da Engenharia de Produção – PPGEP, Universidade Federal de Santa Catarina – UFSC, Florianópolis, SC, Brasil
Abstract
Mood boards (MB) are a visual tool used to provide references for inspiration and decision-making in the design
development process. However, it isn't easy to obtain objective and precise measures about the MB organization if
exclusively qualitative approaches identify how the image composition can affect the visual reading process. In this sense,
this study aimed to evaluate how the position and physical image aspects influence an attentive MB visual reading. The
results presented that variations in image contrast and positioning influence the observer's visual reading. Besides, the
observers' identified tendencies were to concentrate the visualisations in the boards' central, lower and left areas,
indicating the importance of this region's MB construction. Moreover, the results evidence that the visual reading of an
MB presents a different pattern from those reported by studies directed to analysing the visual reading of websites. Thus,
changes in image positioning influenced the viewer's perception, being a relevant point to be considered when assembling
the MB's image composition. The results of this study contribute to creating criteria for the construction of MB.
Keywords: mood boards, design tools, boards composition, perception.
1. Introduction
Mood boards (MB) are visual tools used in various parts of the design development process to convey an
orientation, vision, or idea to different stakeholders (Chang et al., 2014; Lucero, 2012; McDonagh & Denton,
2005). These boards help to serve as a tangible object, a visual reminder to inspire solution principles (Martin &
Hanington, 2012). Made up predominantly of images, MB are considered an easy, quick, and organised way of
visually referring to a specific technical design, concepts alignments or aesthetic attributes. In this context, the
MB presents itself as an important resource for developing a product, making visually technical, aesthetic, and
symbolic elements available to subsidise design decisions (Löbach, 2001).
The MB construction process involves selecting and composing images, colours, lines, textures, words and
objects in a defined space (Cassidy, 2011; Edwards et al., 2009; Lucero, 2015). For its semantic characteristic,
images are an appropriate resource to convey meanings, particularly values and emotional experiences
(Bruseberg et al., 2003). The evaluation of an MB, regarding its content, context and visual appearance, should
be considered an inherent part of the creation process, the decisions made, the reasons for those decisions, and
the reflections undertaken (Cassidy, 2011). Product designers use them to explicitly represent spaces,
movements, surfaces, and materials (Lucero, 2012). In contrast, graphic designers choose images for visual
qualities, such as colours, font characteristics, and textures, to stimulate design associations (Lucero, 2012).
It has already been evidenced that each MB is constituted according to the context needs, with visual elements
positioned and directed to meet different project demands (Baxter, 2011; Gade, 2016). Due to the imagery
resources being increasingly used in different product development stages and the scientific community to
disseminate the results obtained, developing MB guidelines can turn more efficient in transmitting helpful
information for decision-making.
The MB composing activity is associated with the image elements' subjective choices, associating creativity,
personal experiences, and cultural context to the board (Gonçalves et al., 2014; McDonagh & Denton, 2005). For
Edwards et al. (2009), there is no existing methodology for MB composition, being creativity and imagination
the primary resources used in this activity. The MB composition activity starts from personal or group image
choices, using creativity, individual preferences, and cultural context (McDonagh & Denton, 2005), resulting in
unique compositions, visual elements, and the way they are fixed and presented. When composing an MB, several
relevant factors must be taken into consideration, including not only the positions of visual elements within the
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 2/12
board, oriented to aesthetically balanced composition (Arnheim, 2004), but also: their colours (P. Locher et al.,
2005); orientations (Locher & Stappers, 2002); sizes, and shapes (Hübner & Thömmes, 2019; Wilson &
Chatterjee, 2005). A balanced composition points to a tendency to distribute visual elements throughout the
physical space of the board (Leyssen et al., 2012).
MB are commonly observed in effective attention conditions, where there is a selective search for visual
elements, symbols, shapes and colours in the set and details of images. Visual reading sequence, carried out
systematically, plays a central role in how information is perceived and absorbed. Previous studies assessing user
reactions to a range of websites have shown that the location of an element is a critical factor in determining its
place in the viewing order (Djamasbi et al., 2011). Hence, understanding how different observers view an MB
favours better control over its composition, benefiting communication and the reading experience. This
visualisation can be achieved through a targeted selection of elements and attributes, such as size and position
(Faraday, 2000). Thus, the images position or other visual elements can affect the reading process by including
prominent features that attract attention to a particular point of the board.
Developing a study on proper positioning and selecting images for an MB is one of the gaps to be filled by
design professionals (Koch et al., 2020). Moreover, the step-by-step process of building an MB is a process
evidenced by many works (Cassidy, 2011; Lucero, 2015; Rieuf, 2013). However, within the different stages of
building an MB, there are processes, such as positioning and the choice of image, which require more evidence
(Koch et al., 2020).
In this sense, organising an MB should cover aesthetic concerns and technical aspects favouring visual reading
control, making the message more evident and efficient. Baxter (2011) showed that an MB could trigger different
feelings in the users. Therefore, the dynamics of using an image (positioning, framing, dimensions, details) are
substantial factors for the development of an MB, as it is a relevant tool for users to express their emotions,
expectations and views (McDonagh et al., 2002).
Given this, highlighting the influence of positioning, dimensions, format and details of images in MB can
allow a better domain over this construction process, making it more controlled and efficient. In this sense, the
aims of this study evaluated, through an objective approach, how the position and physical image characteristics
influence an attentive MB visual reading. Thus, this study shows a closer understanding of the visual reading
process of an MB composition.
In order to elucidate the above questions, this research was conducted based on separate stages linked to the
selection of observers, the visualisation of different MB with the use of eye-tracking glasses and, finally, the
analysis of the collected data. Thus, the influence that images exert on the visual reading of distinct MB formats
were identified.
2. Mood boards
Godlewsky (2008, p. 266) defines MB as “[...] a collage implemented to introduce a certain mood, theme, or
consumer world”. The collage of previously selected images intent to hold significant details about different
qualities of the desired product. It can also help the team assimilating the mood of a form of creation's values.
MB have two essential elements, a board and a message (Cassidy, 2011). The board can be made of any flat and
unmarked material, which serves as a centralising and aggregating area for a message. The board is a critical part
of the tool and requires a good perception of those who build them to guide the design team correctly. The MB
message is characterised by a set of visual elements previously selected and randomly arranged on the flat board
(McDonagh & Denton, 2005). A poorly constructed MB can drive messages not in tune with the aesthetical
specifications (Garner & McDonagh-Philp, 2001). Pazmino (2015, p. 105) suggests that the board would be “[...]
assembled with images, however, one should consider that a text gives greater emphasis to the characteristics of
the image, not least because the image alone can be misinterpreted”. In this sense, the use of a wide range of
visual elements for the composition of the board is highly encouraged, as long as they are aligned with the
context, such as landscape photographs, small objects, words, artistic drawings (own or others), sketches of
stylistic details, geometric shapes, random lines, scents, textures. Anything that can generate a composition of
elements capable of transmitting a strong message about meaning is appropriate and harmonised on the board.
As for the MB visual organisation, they can be assembled in different configurations (Figure 1): (1) as a
collection of images in which each one represents something (Yamani et al., 2010); (2) as a collection of images
that look like or are related to something (Cassidy, 2008); or (3) as a collection of images, having a central one
and flanked by others that support it, related to something (McDonagh & Denton, 2005). Pereira (2010, p. 39)
comments that “while some professionals choose to separate the board references in aspects to be planned (colour
palette, accessories, environments, for example) and others seek to approximate them to facilitate the expression
of an idea, some designers work with a logic of construction undefined”.
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 3/12
Figure 1. Different configurations of MB visual organisation.
2.1. Visual reading
Visual reading, exploring a specific item or object in a scene, is a ubiquitous part of human visual behaviour (Findlay
& Gilchrist, 2003). It involves more than the contemplative observation that occurs when one unassumingly observes
some landscape. When proceeding with visual reading, an object is analysed and visualised with attention to perceive all
the form nuances and details, its surroundings and the context in which it is located, seeking to understand the messages
and meanings of this set (Collewijn, 1998). This analysis identifies and evaluates the elements that make up an object
under study, including finish, technical information, structural details, and other specifications.
The scanning process for visual information begins in the centre, but the paths of the reading sequence may
vary. Hooge & Erkelens (1998) used circular displays in an experiment and reported that subjects tended to scan
clockwise or counterclockwise around the display. Zelinsky et al. (1997) observed another pattern in an
experiment with a small number of items on a table, that subjects made several saccades to locate the target.
Fixation positions analysis of the saccades during the scanning path showed that they were initially directed
toward the centre of the item groups and progressively focused on smaller details.
Looking is not something passive and, according to Pylyshyn (2003, p. 160), “[...] we affect what we see when
we choose where to look. Indeed, changing our direction of gaze by moving our eyes is one of the main ways in
which we selectively adjust the visual world.” However, there are some common characteristics regarding the way
this occurs. The selective search for elements of interest usually has, as a matter of naturalness, its beginning in the
centre of an image or object (Arnheim, 1982, p. 4). It is the first contact with the visual element, recognising that
there is something to be explored in that place. Next, the eye is attracted to some point of interest and makes a whole
object visual reading. Internet sites, for example, tend to have the upper left corner as its most valued part, for being
the first to receive effective attention from the reader (Damasceno & Gruszynski, 2014; Faraday, 2000). It keeps
some relation with reading texts in occidental culture, which has the same place as the initial point of attention.
According to Sternberg (2010, p. 107), “[...] attention is how a limited amount of information is actively
processed from the enormous amount of information available through the senses, stored memory, and other
cognitive processes”. Attentively observing an object requires a person to focus on a small part of it to absorb
the available information and process it properly. Attention, in turn, allows one to select some of the available
visual information for longer or more detailed analysis (Findlay & Gilchrist, 2003, p. 3).
Attentive visual reading applies to any object. Searching comprises actively and skillfully seeking a target
(Pashler, 1999). When a person reads an MB image, it works with attention, essentially in the search function,
looking for semantic elements matching a personal reference (Sternberg, 2010). The purpose is to select these
elements as a visual reference, team alignment, or stimulus to create concepts and comparisons.
When it comes to design tools, such as MB, attention plays a relevant role for the designer when analysing details of
the images and capturing helpful information for application in the project. By studying the visual reading procedure, one
seeks to increase the efficiency of the tool's application and, consequently, the results provided to the project.
3. Methodological procedure
In order to evaluate the image dynamics in different types of MB, an individual eye-tracking reading capture was
conducted for data collection. This methodology was employed to identify whether the positioning of different images and
the image style can affect the visualisation of an MB. Subsequently, the data were analysed with a statistical basis and thus
presented.
Eye-tracking allowed examining the user's reaction to viewing the MB in real-time. Eye-tracking makes it possible to
record the movement the eyes make when scrolling through an image. It examines the user's reaction to viewing the boards
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 4/12
in real-time, while traditional methods (interviews, focus groups, questionnaires) are flawed because they depend on the
willingness and competence of the participants to describe how they feel when they are exposed to some stimulus
For data collection, participants' choice used, as inclusion criteria, design undergraduate course students at Federal
University of Santa Catarina (southern Brazil), enrolled in Project VI subject, and MB familiarised. Twelve participants
were selected, read and accepted the Informed Consent Form (ICF) terms. According to Pernice & Nielsen (2009), in a
qualitative eye-tracking study, six users are considered enough to assess the main usability issues accurately. After putting
on the eye-tracking goggles, adjustments and calibration of the equipment, the participants were instructed to observe three
different MB for a total time of 120 seconds (3x40 s). The application of the procedure had an average duration of 8
minutes per participant.
The experiment to capture the visual reading through the eye-tracking equipment included preparing two MB
sequences to stimulate and capture the observation profile in each one of them. The difference between the two sequences
is in the MB used, with similar composition and identical images, but with changes in their positioning. In this sense, five
boards were prepared, divided into three types: A, B, and C. Thus, 6 people observed the MB sequence 1, and the other 6
people observed the MB sequence 2. The characteristics of each MB are described in Table 1 and presented in Figure 2.
Figure 3a shows the sequences with the types of MB used in this study.
To study the eye movements in the visual reading process related to each Area Of Interest (AOI), we chose to divide
the boards into nine AOI's (Figure 3b) and to interpret the following visualisation criteria, generated by the BeGaze
software: (i) time to first fixation (time from the start of viewing until the participant fixates gaze for the first time on an
AOI); (ii) average duration of fixations (average duration of all fixations on each AOI); (iii) fixation count (number of
times the participant fixated on an AOI; (iv) revisit (number of times the participant returned to an AOI); (v) dwell time
(time the participant spent looking at an AOI); (vi) sequence (sequence of visual reading of the AOI). The analysis was
also performed with the help of the “heat map” resource to observe the places with higher and lower incidence of
visualisation.
For this study, were used two laptop computers and one SMI Eye-tracking glasses. One laptop equipped with BeGaze
v. 3.6 software for eye-tracking data capture and analysis. Another laptop was used to calibrate the tracking glasses and
present the MB to the participant during the reading capture procedure.
Table 1. Boards used to compose the visualisation sequences.
Boards
Characteristics
Objective
Type A
· A board composed of 13 image elements;
· Obtain data on the visual reading of all participants
to detect patterns, deviations and eventual disparities
in board reading;
· 12 images of
residential interiors, with varied visual
complexity and two primary colours (white and
brown);
· Obtain data to determine if variations in the number
of visual elements than other boards influence visual
reading behaviour.
· Textual elements containing
the words “wood &
white" and "mood”;
· Used, without changes, in both visualisation
sequences.
Type B
· Boards composed of 31 image elements;
· To obtain comparative data to assess whether
objects' position influences the visual reading
sequence.
· 31 small images of decorative and monochrome
(black and white) objects of various sizes;
· Board B2 contains the same images as B1 but with
horizontally inverted positions.
Type C
· Boards composed of 22 image elements;
· To obtain
data indicating whether a visual element
with colour contrast influences visual reading
behaviour.
· 21 Images in intense, artificial colours and cool
tones;
· One colour palette on the left of the board;
· One of the images includes an
object in a contrasting
colour (orange);
· The C2 board contains the same images as C1 but
in randomly switched positions.
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 5/12
Figure 2. Boards used for the visual capture process.
Figure 3. (a) Boards presentation sequence in the visual reading capture process. (b) Letters assigned to the nine MB Areas
of Interest (AOI's).
With the results obtained, for each MB visualisation criterion identified, was evaluated if a statistical
difference occurred (ANOVA analysis of variance - p<0.05): (a) between the same MB sequence (1 x 2); (b)
between different boards of the same sequence (A x B x C); (c) different AOI of the MB (Figure 4). Furthermore,
were conducted Pearson correlations to verify directly or inversely proportional behaviour between the results of
the visualisation criteria. The StatSoft Statistic 12.0 software conducted all statistical analyses.
Figure 4. Statistical analysis procedure
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 6/12
The results are shown, taking into consideration the data provided by the BeGaze software. Subsequently, a
statistical analysis was presented regarding the data collected from the MB and their respective AOI, and finally,
an integrative discussion of the results was conducted.
4. Results
4.1. MB type observer influence
Considering the data collected (Appendix 1), an exciting behaviour was identified with the participants'
responses when different MB was observed (Figure 5). Firstly, different MB showed to influence the response
of the viewers. The performance can be explained because there are no statistical differences (p = > 0.05)
identified when comparing MB A1 and A2, as well as MB B1 and B2. In MB types A and B, there was no
statistically significant difference between them. On the other hand, there was a statistical difference (p= < 0.05)
with random images positions in MB type C. In this case, changes in image positioning have been shown to
influence the observer's perception, being a relevant point to be considered.
Figure 5. Visualization criteria mean values and standard deviation identified for each board. A) Entry time; B) Revisits; C)
Dwell time; D) Average fixation; E) First fixation; F) Fixation count.
Furthermore, a statistical difference (p = > 0.05) was also identified for all MB within the same sequence.
These results evidenced that the image arrangement within an MB is a factor to be considered when constructing
the MB.
Overall, Pearson correlations showed a significant inversely proportional association (-0.875) between
average entry time and average fixation, indicating that the average fixation decreases when entry time increases,
showing that the MB image set influenced these variables. The shortest average time required for the first input
to occur, by the observer, in each AOI, was identified for MB A1 and A2 (average values of 5,044.5 and 3,501.2
ms, respectively) when compared to the other MB (average values range from 5,594.2 to 6,718.0 ms). The longest
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 7/12
average dwell time also occurred for the AI and A2 MB, with 29840.1 ms, compared to the other MB, which has
average values of 24137.4 ms for the B1 and B2 boards and 27584.1 ms for the C1 and C2 boards. This data
seems to be associated with the number of imagery elements of the MB different types, pointing to a proportional
relationship between boards. MB A1 and A2 have a smaller number of image elements than the others, with 13
elements. MB type B has 31 elements, and MB type C has 22 elements. The MB type C was the boards with a
longer time for participants to fix their gaze (average values of 173.3 ms for C1 and C2) than the other MB
(average values range from 151 to 164 ms). This behaviour may be associated with the type of image used, with
colours in exotic shades, generating a greater degree of curiosity and comparative observation between fantasy
and reality.
Pearson correlations are directly proportional (0.9835) between the dwell time, the time the participant spent
looking at an MB AOI, and the number of revisits identified, indicating that when one variable increases, the
other variable tends to increase as well. In this sense, MB A1 and A2 presented the longest dwell time (3259.2
and 3371.9 respectively), while the other MB presented mean values in the range of 2753.7 to 3318.9.
Simultaneously, the number of revisits was higher in A1, A2 and C1, with average values of 4. The other MB
had average values of around 3. A higher number of revisits in the MB may be related to a less attentive reading
by the participant. It indicates a higher number of times that the participant returned to the same AOI. On the
other hand, the longer dwell time may indicate a high level of interest in one or more specific AOI.
Based on the results obtained, the MB's sequence 1 had higher numerical values than sequence 2 for all the
equipment's data. This behaviour showed that the reading performed by the participants of sequence 1 was less
attentive than sequence 2. However, the differences presented suggest that they were related to individual
differences between the participants who made the observations.
4.2. AOI influence on the MB visual reading
To evaluate the influence of AOI on the board visual reading, MB were delimited into 9 areas to compare the
obtained data between them. It can be identified that the AOI that most influenced the results varied for each type
of MB. Statistical differences were observed for all the criteria evaluated between the different AOI of the
different boards (p> 0.05).
A heat map was employed to show the aggregate fixations based on the results obtained (Figure 6). The results
generated indicate a higher observation frequency in the central parts of the images when viewing the imagery
elements individually. Pylyshyn (2003) points out that a variety of evidence suggests that visual attention
operates - at least, in part - on whole objects rather than places or regions or any other spatially defined aspects.
On the other hand, the map also points to a tendency by participants to direct their gaze to the lower third of the
visualised imagery elements.
Figure 6. Heat map identified for each board.
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 8/12
The most significant visual fixation points in the specific AOI of each board presented were highlighted
(Figure 7). In MB type A, they occurred in AOI G. In MB type B and type C, they occurred in AOI E (centre of
the board). The average fixation of all boards indicated a tendency to concentrate fixations in the central and
lower board area. The data also indicate a reading pattern that differs from that observed in webpages, which
point to the upper left area as the one with the highest concentration of visual readings (Faraday, 2000).
Figure 7. Average Dwell time. The red colour graduation indicates higher dwell time values.
Among boards of the same type, for 2/3 of the AOI with the same position, in MB A, presented values in the
fixations are close, suggesting a similar visual behaviour among participants. In MB type B, with inverted
horizontal compositions, an approximation of values between similar AOI was also detected, suggesting that the
simple inversion in the imagery compositions does not reflect a similar behaviour in viewing the board. In MB
type C, which presents a contrast object (orange), an increment in the fixation value was detected when the object
was in the respective AOI. The board C1, with the contrast object in AOI E, the fixation recorded in this AOI
was 17.0%, the highest percentual in the whole fixation data capture, while the value for AOI B was 3.6%. It is
possible to deduce that this 17.0% value obtained in AOI E results from the centre of the board and the contrast
object. The board C2, with the object in AOI B, the fixation recorded was 11.5%, and the value recorded for AOI
E = 14.0%. This change suggests that, even in situations of close observation, the use of contrast elements
influences the visualisation of an MB. Colours can exert a strong emotional and behavioural appeal (Cyr et al.,
2010) and, in this case, may also exert some influence on the visual reading process.
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 9/12
5. Conclusion
Based on an MB visual reading process analysis linked to different viewers, the image position and physical
characteristics that influence an attentive MB visual reading were identified. Thus, it can be concluded that:
Based on the different data identified for different MB types, image positioning influenced the viewer's
perception, which is a relevant point to be considered when assembling the MB's image composition.
The AOI that most influenced the results varied for each MB. However, a tendency was identified on the
participant to concentrate the readings in the central area and the lower and left halves of the boards, overlapping
the characteristics applied between the MB used (inversion in image composition and contrast by colour).
Considering the reading of an MB as an action of attentive reading, in which the observer operates a visual
scan, visual resources, such as the colour contrast presented in boards C, proved effective in creating
differentiated areas of visual attraction. In this sense, contrasting images is a visual resource to be observed when
composing an MB.
Eye-tracking equipment effectively detects visualisation actions, presents AOI data, and compares the
different MB. The time for visual reading of each board (40 s) was adequate for complete visualisation and data
collection. However, in research involving the analysis of average fixations, the exposure time should be
extended to allow participants to perform average fixations between 200 and 300 milliseconds. According to
Granka et al. (2008, p. 348), this is the period “[...] during which visual attention is directed to a specific area of
the visual display”.
As an opportunity for future research, studies can be carried out with other imagery compositions, evaluating
background colours, sizes and orientation of images, geometric elements, objects (on physical boards) and videos
that influence the reading. Research involving the format, location, access and quantity of MB may also improve
its efficiency as a tool.
References
Arnheim, R. (1982). The Power of the Center: a study of composition in the visual arts. Berkeley: University of California Press.
https://doi.org/10.2307/3332229
Arnheim, R. (2004). Art and visual perception: a psychology of the creative eye (2nd ed.). Berkeley: University of California Press.
Baxter, M. (2011). Projeto de Produto: guia prático para o design de novos produtos (3rd ed.). São Paulo: Edgard Blücher.
Bruseberg, A., McDonagh, D., & Wormald, P. (2003). The use of images to elicit user needs for the design of playground
equipment. In D. McDonagh, P. Hekkert, J. van Erp, & D. Gyi (Eds.), Design and emotion (pp. 114–118). CRC Press.
https://doi.org/10.1201/9780203608173-c21
Cassidy, T. (2008). Mood boards: current practice in learning and teaching strategies and students’ understanding of the process.
International Journal of Fashion Design, Technology and Education, 1(1), 43-54. http://dx.doi.org/10.1080/17543260802015154.
Cassidy, T. (2011). The mood board process modeled and understood as a qualitative design research tool. Fashion Practice, 3(2),
225-251. http://dx.doi.org/10.2752/175693811X13080607764854.
Chang, H.-M., Díaz, M., Català, A., Chen, W., & Rauterberg, M. (2014). Mood boards as a universal tool for investigating
emotional experience. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and
Lecture Notes in Bioinformatics): Vol. 8520 LNCS (Pt 4, pp. 220–231). Springer. https://doi.org/10.1007/978-3-319-07638-6_22.
Collewijn, H. (1998). Eye movement recording. In R. Carpenter & J. Robson (Eds.), Vison research: a practical guide to laboratory
methods (pp. 245-285). Oxford: Oxford University Press.
Cyr, D., Head, M., & Larios, H. (2010). Colour appeal in website design within and across cultures: a multi-method evaluation.
International Journal of Human-Computer Studies, 68(1-2), 1-21. http://dx.doi.org/10.1016/j.ijhcs.2009.08.005.
Damasceno, P. L., & Gruszynski, A. (2014). Design de jornais - processos, rotinas e produto: um estudo do Segundo Caderno,
suplemento cultural de Zero Hora. Brazilian Journalism Research, 10(1), 108-127. Retrieved in 2021, September 3,
from https://bjr.sbpjor.org.br/bjr/article/viewFile/551/530
Djamasbi, S., Siegel, M., Skorinko, J., & Tullis, T. (2011). Online viewing and aesthetic preferences of generation y and the baby
boom generation: testing user web site experience through eye tracking. International Journal of Electronic Commerce, 15(4), 121-
158. http://dx.doi.org/10.2753/JEC1086-4415150404.
Edwards, A., Fadzli, S. A., & Setchi, R. (2009). A comparative study of developing physical and digital mood boards. Proceeding
Innovative Production Machines and Systems, 1-6.
Faraday, P. (2000). Visually critiquing web pages. In N. Correia, T. Chambel & G. Davenport (Eds.), Multimedia ’99.
Eurographics. Vienna: Springer. https://doi.org/10.1007/978-3-7091-6771-7_17.
Findlay, J. M., & Gilchrist, I. D. (2003). Active Vision: the psychology of looking and seeing. Oxford: Oxford University Press.
https://doi.org/10.1093/acprof:oso/9780198524793.001.0001
Gade, U. T. (2016). Design boards as an alignment tool for cross-disciplinarity in engineering. In E. Bohemia, L. Buck, K. Eriksen,
A. Kovacevic, N. Ovesen & C. Tollestrup (Eds.), Proceedings of the 18th International Conference on Engineering and Product
Design Education: Design Education: Collaboration and Cross-Disciplinarity, E and PDE 2016. Institution of Engineering
Designers, The Design Society.
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 10/12
Garner, S., & McDonagh-Philp, D. (2001). Problem Interpretation and Resolution via Visual Stimuli: The Use of ‘Mood Boards’ in
Design Education. International Journal of Art & Design Education, 20(1), 57-64. http://dx.doi.org/10.1111/1468-5949.00250.
Godlewsky, T. (2008). Mood board. In M. Erlhoff & T. Marshall (Eds.), Design dictionary: perspectives on design terminology (pp.
266). Birkhäuser Basel. https://doi.org/10.1007/978-3-7643-8140-0_173.
Gonçalves, M., Cardoso, C., & Badke-schaub, P. (2014). What inspires designers? Preferences on inspirational approaches during
idea generation. Design Studies, 35(1), 29-53. http://dx.doi.org/10.1016/j.destud.2013.09.001.
Granka, L. A., Feusner, M., & Lorigo, L. (2008). Eye monitoring in online search. In R. I. Hammoud (Ed.), Passive Eye
Monitoring: algorithms, application and experiments (pp. 347-372). Springer-Verlag. https://doi.org/10.1007/978-3-540-75412-1
Hooge, I. T. C., & Erkelens, C. J. (1998). Adjustment of fixation duration in visual search. Investigative Ophthalmology & Visual
Science, 38(4), 1295-1302.
Hübner, R., & Thömmes, K. (2019). Symmetry and balance as factors of aesthetic appreciation: Ethel Puffer’s (1903) ‘Studies in
Symmetry’ revised. Symmetry, 11(12), http://dx.doi.org/10.3390/SYM11121468.
Koch, J., Taffin, N., Lucero, A., & Mackay, W. E. (2020). SemanticCollage: enriching digital mood board design with semantic
labels. DIS 2020 - Proceedings of the 2020 ACM Designing Interactive Systems Conference, 407-418.
https://doi.org/10.1145/3357236.3395494.
Leyssen, M. H. R., Linsen, S., Sammartino, J., & Palmer, S. E. (2012). Aesthetic preference for spatial composition in multiobject
pictures. I-Perception, 3(1), 25-49. http://dx.doi.org/10.1068/i0458aap.
Löbach, B. (2001). Design Industrial: bases para a configuração dos produtos industriais. São Paulo: Edgard Blücher.
Locher, P. J., & Stappers, P. J. (2002). Factors contributing to the implicit dynamic quality of static abstract designs. Perception,
31(9), 1093-1107. http://dx.doi.org/10.1068/p3299.
Locher, P., Overbeeke, K., & Stappers, P. J. (2005). Spatial balance of color triads in the abstract art of Piet Mondrian. Perception,
34(2), 169-189. http://dx.doi.org/10.1068/p5033.
Lucero, A. (2015). Funky-design-spaces: interactive environments for creativity inspired by observing designers making mood
boards. In J. Abascal, S. Barbosa, M. Fetter, T. Gross, P. Palanque, & M. Winckler (Eds.), Lecture Notes in Computer Science
(including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9298, pp. 474–492).
Springer International Publishing. https://doi.org/10.1007/978-3-319-22698-9_32
Lucero, A. (2012). Framing, aligning, paradoxing, abstracting, and directing: how design mood boards work. In Proceedings of the
Designing Interactive Systems Conference, DIS ’12 pp. (438-447). https://doi.org/10.1145/2317956.2318021.
Martin, B., & Hanington, B. (2012). Universal methods of design: 100 ways to research complex problems, develop innovative
ideas, and design effective solutions. USA: Rockport Publishers. https://doi.org/10.5860/CHOICE.49-5403.
McDonagh, D., Bruseberg, A., & Haslam, C. (2002). Visual product evaluation: exploring users’ emotional relationships with
products. Applied Ergonomics, 33(3), 231-240. http://dx.doi.org/10.1016/S0003-6870(02)00008-X.
McDonagh, D., & Denton, H. (2005). Exploring the degree to which individual students share a common perception of specific
mood boards: observations relating to teaching, learning and team-based design. Design Studies, 26(1), 35-53.
http://dx.doi.org/10.1016/j.destud.2004.05.008.
Pashler, H. E. (1999). The psychobgy of attention. Cambridge: MIT Press.
Pazmino, A. V. (2015). Como Se Cria: 40 métodos para design de produtos. São Paulo: Edgard Blücher.
Pereira, T. V. (2010). Mood Board como espaço de construção de metáforas. UNISINOS. Retrieved in 2021, September 3,
from http://www.repositorio.jesuita.org.br/handle/UNISINOS/3029.
Pernice, G., & Nielsen, J. (2009). How to Conduct Eyetracking Studies. Fremont: Nielson Norman Group.
Pylyshyn, Z. W. (2003). Seeing and Visualizing: it’s not what you think. London: Bradford Book.
Rieuf, V. (2013). Impact of the immersive experience on kansei during early design [Arts et Métiers ParisTech]. Retrieved in 2021,
September 3,
from http://www.academia.edu/download/32204012/Impact_of_the_Immersive_Experience_on_Kansei_During_Early_Immersive_
-_Vincent_RIEUF_-_LCPI.pdf
Sternberg, R. J. (2010). Psicologia cognitiva (5th ed.). Boston: Cengage Learning.
Wilson, A., & Chatterjee, A. (2005). The assessment of preference for balance: introducing a new test. Empirical Studies of the Arts,
23(2), 165-180. http://dx.doi.org/10.2190/b1lr-mvf3-f36x-xr64.
Yamani, Y., McCarley, J. S., & McDonagh, D. (2010). Transgenerational communication through affective imagery in mood
boards. Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 54(20), 1762-1765.
http://dx.doi.org/10.1177/154193121005402004.
Zelinsky, G. J., Rao, R. P. N., Hayhoe, M. M., & Ballard, D. H. (1997). Eye movements reveal the spatiotemporal dynamics of
visual search. Psychological Science, 8(6), 448-453.
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 11/12
Appendix 1. Data provided by BeGaze software.
Board AOI Sequence
Entry time
(ms)
Dwell time
(ms)
Revisits
Average
fixation (ms)
First fixation
(ms)
Fixation
count
A1
A 2 2576.9 2704.5 3.0 143.8 210.6 6.3
B 7 7453.7 2031.1 2.9 129.5 205.1 6.2
C 9 8224.0 2865.1 3.0 219.6 188.4 6.2
D 1 2327.5 4028.9 6.0 137.5 166.2 12.5
E 4 3882.0 2430.1 2.9 84.7 205.1 6.2
F 6 5419.8 1945.2 4.2 141.8 166.2 9.5
G 8 7752.9 6422.8 4.7 214.3 205.0 17.7
H 3 3391.6 2172.4 3.9 157.5 310.3 8.7
I 5 4372.4 4732.9 5.2 235.5 160.7 14.7
Average
5044.5 3259.2 4.0 162.7 202.0 9.8
A2
A 7 5287.0 2493.9 3.0 126.7 177.3 6.0
B 8 1490.8 3480.3 3.5 131.1 177.4 10.0
C 9 8733.9 2848.5 3.2 243.5 238.3 6.8
D 4 2621.3 4450.1 6.3 149.1 171.8 13.5
E 2 1490.8 3670.6 3.5 84.4 155.2 10.0
F 1 1052.9 1767.8 2.0 175.3 138.6 7.8
G 5 4234.0 6062.8 5.0 214.3 182.9 16.7
H 3 1712.4 2154.0 4.5 146.5 155.2 8.8
I 6 4887.9 3419.3 4.8 165.9 155.2 11.7
Average
3501.2 3371.9 4.0 159.6 172.4 10.1
B1
A 5 6395.2 2222.3 2.3 178.7 127.5 6.0
B 7 8916.8 1662.5 1.7 154.6 61.0 4.0
C 6 7947.0 1715.1 1.4 120.0 99.7 2.7
D 9 10080.6 3862.8 3.8 201.6 155.2 7.2
E 2 3009.2 4680.1 6.2 140.9 110.8 9.7
F 1 2139.1 4134.3 4.2 179.3 116.4 6.7
G 4 6324.3 1241.3 2 173.4 133.0 3.5
H 3 5658.2 1834.4 3.5 141.1 66.5 5.7
I 8 9991.9 3424.9 3.7 186.6 127.5 7.0
Average
6718.0 2753.1 3.2 164.0 110.8 5.8
B2
A 2 1895.3 3241.9 2.4 167.6 149.6 5.7
B 8 10108.3 1665.3 2.6 118.0 127.5 4.5
C 9 12951.2 1898.1 1 122.3 88.7 2
D 1 1582.7 3017.4 4.2 141.2 127.4 8
E 4 2200.1 3921.0 4.5 152.0 83.1 6.2
F 5 6350.9 3447.0 3.5 137.8 99.8 7
G 6 7675.4 1435.4 2.3 143.1 110.8 4.5
H 3 2100.3 2532.6 4.5 188.8 127.5 6.7
I 7 9870.0 2338.6 2.7 187.9 177.3 4.7
Average
6081.6 2610.8 3.1 151.0 121.3 5.5
C1
A 7 7867.2 1080.7 2.0 141.5 127.4 3.0
B 9 12045.2 1452.0 1.6 141.3 99.7 2.5
C 8 9886.6 2238.9 2.2 179.5 149.6 5.2
D 3 3408.2 4123.0 5.7 163.3 127.5 16.0
E 1 116.4 7010.4 10.3 201.7 166.3 17.2
F 2 2416.2 2815.2 5.3 187.3 133.0 10.2
G 4 3602.2 1745.7 2.8 182.6 177.4 7.7
H 6 6073.7 5104.0 3.8 175.2 121.8 7.7
I 5 4932.2 4300.3 4.2 187.8 138.6 10.7
Average
5594.2 3318.9 4.2 173.3 137.9 8.9
Image composition influences on the mood board visual reading process through eye-tracking
Product, Management & Development, 19(1), e20210010, 2021 12/12
Board AOI Sequence
Entry time
(ms)
Dwell time
(ms)
Revisits
Average
fixation (ms)
First fixation
(ms)
Fixation
count
C2
A 8 9310.2 1657.0 2.8 178.6 166.3 8.3
B 3 5148.3 4633.0 5.2 189.3 116.4 12.8
C 5 7032.6 1684.7 2.5 173.4 182.9 5.5
D 2 4993.2 3302.8 3.0 143.6 44.3 6.3
E 1 493.2 5564.2 8.0 188.7 149.6 11.2
F 4 6711.2 2532.5 2.5 195.3 199.5 6.7
G 7 7520.3 2438.4 3.2 173.5 121.9 5.5
H 9 11161.2 2233.4 2.3 175.9 149.6 6.3
I 6 7354.0 1452.0 2.2 141.9 144.1 4.5
Average
6636.0 2833.1 3.5 173.3 141.6 7.5
Appendix 1. Continued...
