Low vision users in graphical user interface interaction : Examining the effects of visual perception parameters on quality of experience through a display style proposal

© Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  1  Low vision users in graphical user interface interaction: Examining the effects of visual perception parameters on quality of experience through a display style proposal. Aydin M. F., Industrial Design Department, Izmir Institute of Technology, Izmir, Turkey, ORCID 0000-0002-3529-5588, [email protected] Yavuzcan H. G., Industrial Design Department, Gazi University, Ankara, Turkey, ORCID 0000-0001-8560-7845, gyavu[email protected] Received: 2024-04-13 | Accepted: 2025-01-30 | Publication: 2025-05-11 Abstract: Interaction design is the act of designing the dialogue between people and systems, services or products. User interface (UI) facilitates sensory and emotional interactions by acting as a bridge between users and products while graphical user interface (GUI) refers to graphical or visual presentations of information The accessibility of GUIs directly impacts the quality of experience (QoE) since the sense of sight plays a pivotal role especially for low vision users throughout the interaction process. Though low vision users can use GUIs on their own, they often face challenges that hinder their QoE. Although existing research has explored accessible UI design principles, studies specifically addressing the accessibility issues of low vision users in GUI interactions remain limited. The goal of this research is to evaluate the effects of visual perception differences between people with low and normal vision on home appliances’ GUI interactions and discuss how these differences affect the QoE of low vision users. To this end, we created a total of 12 washing machine GUI cases based on the GUI model of the best-selling washing machine in Turkey in 2021 and tested these cases online with two groups of 7 participants each with low vision and normal vision. The results infer that low vision participants have accessibility issues on GUIs in terms of colour contrasts, text sizes, display options and control panel distances. The study suggests that using an uppercase larger font (22pt) for just the first syllable of the text in text-based displays can improve accessibility of GUIs for low vision users compared to using the same font size (16pt) in sentence case for all text. Overall, this study contributes to a better understanding of the challenges faced by low vision users in GUI interactions and offers practical recommendations for creating more accessible GUIs through a display style proposal.1 Keywords: Universal Graphical User Interface, Visual Perception, User-product Interaction, Dısplay Accessibility, User Experience Design, Low Vision Accessibility 1. Introduction The evolution of technology has led to a significant change in traditional physical consumer products, resulting in innovative visual, auditory and kinaesthetic product interactions that improve the quality of life. These innovations, adding more functions and features to the 1 This study is a part of the thesis entitled “Investigation of visual perception parameters in the interface interactions of low vision users on home electronics products” submitted to Gazi University Graduate School of Natural and Applied Sciences to fulfil the degree of PhD in industrial design. © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  2  consumer products to meet the needs of a wider range of users, inevitably reveal increased complexity with a reduced accessible use (Lee, 2021). Home appliances, having an important role for people to meet their basic needs and live independently (Lee et al., 2019), are considered as consumer products where disabled individuals have accessibility problems (Lee, 2021). World Health Organization (WHO, 2011) estimates that over one billion people fall into disability category who are alien to use these products. Meanwhile, a pilot interview with a visually impaired teacher in a secondary school for the visually impaired revealed significant challenges in accessibility of home appliances for low vision individuals, especially during the pandemic. The interview results emphasized the increased reliance on home appliances without assistance due to social distancing measures. Although visually impaired users can operate these devices independently, quality of their use experience is often inadequate, especially for products with various analogue and digital interaction elements, such as washing machines. These findings also coincide with the experiences of one of the researchers of this study, a product designer with ten percent vision. According to the researcher’s own experiences, accessibility issues of low vision users when interacting with home appliances can lead to, if not psychological problems, at least loss of time or extra physical effort; that is, a loss of users’ product experience quality. In fact, the problem is that, people with low vision are highly dependent to visual information even if the information is not clear to them (Lee, 2021). Interaction design is the core area for the accessibility of any manmade artefact. It is defined as creating physical and emotional interactions between a person and a product, system, or service. (Kolko, 2010; Morshedzadeh, Ono, & Watanabe, 2016). It is the act of crafting user experience (UX) that enhances and expands the ways people interact, communicate, and work (Preece, Sharp & Rogers, 2015). UX covers all aspects of user-product interaction, including product’s usability, physical feel, emotional impact, functionality, and contextual relevance (Alben, 1996). It is a subjective and often implicit synthesis of personal factors such as interest, effort, satisfaction, desire, cognition and perception, as well as design domains like purpose, method, function, form and structure. The engagement and success of all these aspects represent the users’ quality of experience (QoE) (Alben, 1996) throughout the interaction process. User interface (UI) refers to the bridges where users access to and interact with designs. A UI is evaluated within three areas: physical UI (PUI), logical UI (LUI) and graphical UI (GUI) (Jin & Ji, 2010; Lee, Jin & Ji, 2011; Mendez & Mendoza, 2013). A PUI represents the interaction tools by which a user executes a task physically. The interface concerning the information-specific contents and structures refers to LUI. Finally, a GUI is the interface which covers the presentation of information by visual or graphical items. Vanderheiden and Vanderheiden (1992) examines UI accessibility in five main groups: Display elements, control elements, manipulations, documentation and safety. Display elements represent devices that convey information to the user while control elements are the tools that enable interaction between user and machine. A GUI can act as both a control and display element at the same time. Although existing research has developed standards and guidelines that can increase the potential of home appliances’ GUIs to be more accessible to everyone, they are limited in several ways. For instance, accessibility standards for disabled people by Americans with Disabilities Act (ADA) (US Department of Justice, 2010) are limited to mobility of wheel chaired population. In addition, most ISO standards except for ISO 22411 and ISO 9241-20 omits elderly people and © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  3  many of them such as ISO 9241-20, ISO TR 29138-1, and ISO TS 16071 address web and softwarebased accessibility issues (Lee et al., 2021). Designing “for those with specific needs” (Pullin, 2009) requires empathy and it can lead to misconceptions driven by misinterpretation of those users’ experiences when empathy lacks Goodman, Langdon, & Clarkson, 2007; Lee et al., 2021; Segelström, 2009). Because not all stakeholders who design and produce GUIs have the pre-built empathy ability to apply the given standards (Lee et al., 2021). In fact, studies about the accessibility of GUIs mostly focus on web and software-based contexts because GUIs have a vital role in human-computer interaction (HCI) especially for bridging the information output and user input visually. However, there are several authors outline basic and shared principles for universal accessibility of display and control elements of products. For instance, Vanderheiden and Vanderheiden (1992) suggest that all important visual information should be conveyed via the auditory and/or tactile senses and sequential tasks should be minimized or there should be clear cues about the order. Displays with values that must be observed together should be in the same horizontal or vertical alignment (Cushman and Rosenberg, 1991; Pheasant and Haslegrave, 2005). Moreover, controls must be operable with minimum force, speed, and accuracy and must be designed with the needs of the lowest-capacity user in mind (Cushman and Rosenberg, 1991; Damon, Stoudt, and McFarland, 1966; Vanderheiden and Vanderheiden, 1992). For simplicity, the number of control elements should be as few as possible (Cushman and Rosenberg, 1991; Damon et al., 1966; Vanderheiden and Vanderheiden, 1992). No complex or unnecessary operations should be required to operate the machine (Bridger, 1995). However, these studies are too general for improved accessibility of GUIs and mostly lack of low vision user-specific guides. The aim of this study is to uncover datasets associated with QoE of low vision users in home appliances’ GUI interactions. To this end, we determined the GUI of the best-selling washing machine in Turkey in 2021 as the case to analyse the effects of various display options on QoE of low vision users. First, we modelled a 1/1 scale copy of the light backgrounded version of this GUI as a vector. Then, based on this model, we modelled two different alternatives that included the display features frequently encountered in other washing machine GUIs on the market. Finally, we modelled a new GUI proposal that we developed based on the experiences of one of the authors, the low vision designer. We determined the dial indicators as the basic variable in all these four GUI cases. In order to test colour variables of dial indicators, by taking the sample GUI colours as reference, we created a dark backgrounded copy for each case. In addition, we developed two more alternative dials to test the dial types. We created a total of four cases by producing an alternative for each of these dials, again based on the colours of the sample GUI. As a result, we subjected these 12 cases to an online simulation test with seven low vision and seven normal vision participants. We recorded the task completion times in seconds and errors made by each participant, where they completed a total of 72 tasks. We also tried to reach qualitative data by conducting semi-structured open-ended short interviews with each participant at the end of the simulation test. We analysed the factors affecting the QoE of low vision participants in washing machine GUIs by examining both the qualitative and quantitative data together. The current research is of modern importance and relevance as it contributes to a better understanding of the challenges faced by low vision users interacting with washing machine GUIs and offers practical recommendations for creating more inclusive and accessible UIs. © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  4  2. Visual perception and visual impairment Visual impairment term is used to describe all kinds of sight loss which covers moderate sight loss, severe sight loss and blindness (Sardegna & Shelly, 2002). When researching with visually impaired users, it is important to recognize and understand the level of the impairment. This includes determining whether a particular user is completely blind, functionally-blind or partiallysighted (Jacko & Sears, 1998). WHO (2004) defines visually-impaired people as those with low vision, and blindness with less than 6/18, but 3/60 or more or a corresponding loss of vision. This can correspond to more than 20 degrees of divergence with the best possible eye. Completely-blind people have no light perception or usable vision. People who are functionally-blind can perceive light, but any visionenhancing device (optical aids such as binoculars, magnifiers, and telescopes) can give them more vision. Partially-sighted people have some useful vision features and can use visual enhancement techniques and devices to function alike fully-sighted people (Kraut & McCabe, 1994). An assessment of the visual limitations includes a baseline data for visual acuity, contrast sensitivity, field of view and colour perception (Jacko et al., 1999). Visual acuity refers to a person's ability to resolve the fine spatial details (Kline & Schieber, 1985) from both close and far locations (Dini et al., 2007). Contrast sensitivity tests a person's ability to perceive the pattern stimuli at low to moderate contrast levels. It is the capacity to detect the differences between the adjacent elements (Dini et al., 2007, p. 16). Contrast sensitivity function represents the spatial discrimination capabilities of the visual channels in a comprehensive manner (Wood & Troutbeck, 1994). Useful field of view is the total area where an effective vision is maintained based on the fixed straight edge fixation point (Kline & Schieber, 1985). The size of the field of view is the size of the space that can be embraced at a glance (Dini et al., 2007). A person's ability to distinguish and define the colours is called colour vision (Dini et al., 2007). Additionally, light sensitivity is the eye reaction to light and light changes (Dini et al., 2007). The ability to perceive depth, known as stereopsis and binocularity, is achieved by utilising the variations in the viewing angles of the eyes (Dini et al., 2007). By combining visual information from both eyes, tasks such as accurately judging distances, grasping objects with precision, and gaining a more comprehensive understanding of the visual world can be carried out (Dini et al., 2007). The term motility refers to the ability of the eyes to make coordinated and continuous movements (Corn, 1983). It encompasses the regulation and synchronisation of eye movements, enabling individuals to track moving objects, shift their focus between different points, and sustain stable and synchronised motions. Impaired motility can result in challenges when engaging in tasks that necessitate precise eye movements and coordination. 3. Methods The research was conducted based on the analyses of both qualitative and quantitative data obtained from seven low vision participants with a visual acuity of 0,10-0,40 and were able to use their personal computers and another seven with normal vision (see Appendices). Four of normal vision participants were industrial designers by whom we could get professional suggestions on GUI test cases. We developed an online simulation platform to test the visual qualities of various washing machine GUIs. We created 12 distinct cases by modelling the GUI of a bestselling brand’s washing machine in Turkey in 2021. The tests were carried out between 12th and 30th January 2022 via an online meeting software that allows screen sharing. © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  5  Figure 1. Sample UI Model 3.1. Design of GUI Cases First, we defined top five brands that have the largest market shares of the Turkish white goods’ sector based on the 2019 report by Euromonitor (Özden, Seheri, & Ersan, 2019). Then we obtained best-selling washing machine models of these top five companies in 2021 from two well-known price comparison websites in Turkey. Notably, all these five models share a similar type of UI, featuring a black digital display with a program selection dial on the left and other button adjustments on the right. For the purpose of this study, we defined the GUI shown Figure 1 for testing, given that this brand holds the highest market share. The current GUI models available in the market are presented in the form of dial options, where symbols, texts, and combinations of both are displayed in various sequences and graphic formats. In the provided GUI sample, the display formats for the screen and the options controlled by the buttons primarily rely on textual information. Consequently, the dial GUI offers greater potential for creating diverse graphic display formats compared to the right panel. Hence, it was deemed suitable to experiment with different representations of the dial options in regards to the research goals. During the design phase, we modelled the 1/1 scale of the sample GUI in vectors. Accordingly, the preferred font used in the sample GUI was determined as Helvetica Neue LT Pro Lt with 16point size. In addition, the sample GUI icons exhibited varying dimensions, with a minimum size of 6x6 mm and a maximum length of 6.5 mm. We developed four distinct GUIs those which included the sample GUI model (C1), two design adaptations tailored to specific functionalities (C3, C6), and a design proposal (C5) based on the experiences of low vision designer-researcher of this study. In addition to the original version which featured a light background, we created an identical set of the four GUIs with a dark background (C2, C4, C7, C8) (see Figure 2). Hence we should evaluate the impact of these two variables on the characteristics of the dial colours. The location of the dial indicators was varied in each GUI to prevent participants from habituating to the indicator locations. In cases C1 and C2, we matched all GUI attributes exactly with the sample GUI. However, in C3, C4, C6 and C7, we produced variables by preserving the sample GUI's icon sizes, font and text size. Of these, we obtained cases C4 and C6 by adhering to a popular format frequently encountered in the market. We determined the icons in C3 and C7 through our research on washing machine program icons. © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  6  Figure 2. Dial indicator test cases Text size is one of the most common problems faced by low vision users. Drummond et al. (2004) conducted a study where they tested unmodified Arial font samples of various sizes with 180 participants. They found that using a font size of 16 points is preferable for users with a visual acuity of 6/24, 18 points for those with 6/36, and 22 points for individuals with 6/60 visual acuity. The sample GUI utilized a 16-point Helvetica Neue LT Pro Lt font, which addressed some of these situations but may still pose accessibility challenges in other cases. To address this, we developed a different display model based on the empathy of one of the researchers, a professional visually impaired designer. In this design proposal seen in cases C5 and C8 in Figure 2, the first or first two syllables of the display text that may evoke the target task, or the entire text in cases where there is enough space to position the text, were displayed in a 22-point font and capital letters, and the syllable(s) following these situations, if any, were displayed in a 11-point font and lowercase letters. To determine the correlation between the dial colour and the background colour, we conducted an investigation using the GUI of the sample models (C1 and C2). As part of the test, we designed a dial in a lighter shade, as shown in Figure 3 (C9 and C10). Figure 3. Dial type test cases The dial of the GUI sample is a dark-coloured model where its marker should have twodimensional clarity problems due to the colour contrast. Although the dial of sample GUI had a tactile depth, it should be difficult for low vision users to perceive it from afar. To assess this situation, we added a dial with a light-coloured arrow marker to the simulation (see Figure 3; C11 and C12). 3.2. Simulation test procedures Prior to conducting the tests, we subjected the GUI to a preliminary test to familiarize the participants with the simulation platform (see Figure 4). At this stage we used a null GUI that have number indicators rather than washing programs. Additionally, each participant was asked to adjust the width of the GUI they saw on the internet browser screen to 40 cm (1/1 scale of the © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  7  sample model width) using the browser's zoom feature and a ruler. To maintain accuracy, we monitored the width adjustments of the participant GUIs displayed on our monitors simultaneously with the participants. We informed the participants when these dimensions reached the fixed width of the participant GUI width we determined before the test on our own monitor, so that each participant tested GUIs with the same width. Hence, we compensated for differences in participants' screen resolutions by preparing the GUI simulations in jpeg format that is fixed to the same width. Differences in contrast and colour settings on the participants' monitors were ignored, assuming that the participants perceived them best on their own computers. This was additionally sufficient for us to have each participant test each case using the same monitor settings. Following this phase, under the guidance of the researcher, the participants were randomly assigned tasks to perform on the null GUI. Figure 4. Simulation GUI: (a) null GUI, (b) active buttons and their displays All the buttons located on the right control panel of the test simulation screen were configured to be active. The purpose of the tasks associated with temperature settings, spin preferences, and pre-wash status was to divert the participants' attention towards the right control panel, thereby preventing them from becoming acquainted with the dial indicators. We aimed to assess the participants' level of awareness regarding the dial task they were engaged in by this approach. Moreover, we recorded the number of clicks performed during button tasks to evaluate the participants' proficiency in completing the assigned task. Furthermore, in these supplementary tasks, we enabled all the values within all programs to prevent participants from detecting their errors in the program-related tasks since in reality button adjustments of the washing programs are limited in regards to selected washing programs. Dial routing interaction was provided by discrete “turn right-turn left” buttons supported by 22 point Helvetica Neue LT Pro L font and arrow indicator in 34x34 mm active click area. The reason for choosing such a discrete interaction was to prevent participants' task concentration on the dial from being distracted by platform button interactions. It was foreseen that an interaction to be positioned around or on the dial could affect the performances of the participants and increase © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  8  the margin of errors. In the test simulation, no visual or audial feedback was provided by interactive buttons but the visual change on the display screen and dial itself. The cases were tested in a random sequence to avoid the participants from learning the program display and the logic behind the task queries. In each case, all participants were asked to adjust six programs in different tasks. We aimed to render the dial indicators within a total of 24 steps; 17 on the right and 7 on the left. The tasks were asked by the researcher via vocal input and the mouse moves and clicks of each participant were monitored in a step-by-step fashion. For instance, in case C2 (see Figure 5), the participants were asked to find “perde (curtain)” mode (+5 steps) before they have seen the upcoming GUI. When they were sure that they found the program, they were asked to repeat the task “perde” or say “yes/ok”. Right after the participants’ vocal input, the researcher posed the next query of the task regarding temperature, spin and prewash (e. g. 40 degrees, 800 rpm, with pre-wash) adjustments. Right before the participants’ push to the “başla/beklet (start/pause)” button, the researcher raised the next task query as “pamuklu (cotton)” (+4 steps). The steps were repeated until the completion of six tasks for each case. Figure 5. Video conference screenshot. After the completion of the 12 cases with 72 tasks, we showed the images of the dial indicator (C1-C8) and the dial type (C1, C2, C9-C12) cases respectively to each participant in two separate stages. In these phases, we conducted short open-ended interviews about the participants' experiences with GUIs and dial features within the contexts: • Comfort: Which GUI layout they feel comfortable using and why. • Accessibility: Which dial and ground contrast they can perceive more easily. • Easiness: Which dial type is easier to use and why. • Suggestion: What can be done for a more accessible GUI design in these cases. We asked the participants to evaluate the GUIs and score 1 as the lowest and 5 as the highest. We did not restrict the participants in scoring value repetitions so that they could examine their experiences within a wider perspective rather than grading the GUIs with distinct scores. 3.3. Data collection and analysis procedures We recorded data in video format with the participants’ permissions and analysed the execution times of the 72 tasks in seconds manually for each participant. It was deemed sufficient and appropriate to receive the data in seconds because the participants exhibited very different © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  9  behaviours while completing the tasks, and these differences required participant-specific followups. For example, some participants preferred to use their mouse and take action at the same time when thinking about the tasks but others preferred to find the desired program setting first and then take action. Similarly, some participants completed the dial tasks by constantly searching from left to right, while others did the opposite. Therefore, the GUI tests were administered by the researcher following the cursor movements, button click behaviours, and voice feedbacks for each participant rather than analysing the number and duration of mouse clicks with a software that may cause data deviations. 4. Results 4.1. Dial indicator test results Throughout the analysis, we did not observe any discrepancies in the time required to complete the dial and button tasks. However, 6 out of the 7 low vision participants were unable to successfully complete cases C3, C4, C6, and C7, resulting in a total of 64 errors, (see Table 1). This was attributed to the fact that the simulation presented the same choices for each mode of button adjustments, making it difficult for participants to identify their mistakes. Consequently, the participants spent more time on both dial and button adjustments for these particular tasks due to the confusion they experienced. Table 1. Low vision users' completion times and total errors in dial indicator tasks (C1-8). Cases L1 L2 L3 L4 L5 L6 L7 Avg. time (sec.) Total errors C1 25 101 78 46 45 39 63 57 1 C2 27 60 47 54 46 44 59 47 - C3 48 88 87 49 70 76 78 71 16 C4 36 118 53 40 55 63 82 64 18 C5 27 63 52 40 38 41 56 45 - C6 28 84 83 44 64 81 95 68 15 C7 29 103 57 38 62 96 83 67 14 C8 25 56 39 40 40 37 51 41 - 57,5 64 L: Low vision user Based on the Skewness and Kurtosis values of C1-C8 tasks for low vision participants falling within the range of +2 and -2, the test times of each case showed a normal distribution. However, the homogeneity test exhibited a Sig.(p) value of 0.037<0.05 indicating that the variances were not distributed uniformly. Anova Welch Test also showed a significant difference between the test times of low vision participants in C1-C8 cases with a Sig.(p) value of 0.02<0.05. Post Hoc Test results revealed that only the C3 and C8 cases have a significant difference with a Sig.(p) value of 0.032<0.05. The mean difference values (29,714) suggested that this difference was in favour of C8. Normal vision users' time scores and errors made in C1-C8 tasks are shown in Table 2. Since the test of normality for Skewness and Kurtosis values of the cases were between +2 and -2 values, © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  16  in the same font size. Furthermore, we suggest that a text display in search-and-find tasks can be perceived more easily by low vision users, regardless of the size of the rest of the text, if the first syllable or the first part which evocates the meaning of the entire text is written in uppercase letters and a larger font. The study’s holistic approach, integrating quantitative data, qualitative assessments, and user feedback, provides a detailed understanding of the complex interplay between design elements and QoE in GUIs, with implications for both low vision and normal vision users. Ultimately, these findings can inform future developments in GUI design, promoting inclusivity and accessibility across diverse user groups. 7. Acknowledgements We would like to thank Prof. Dr. Onder Erkarslan and MSc. Ferda Gul Aydin Emekligil for their support on this article. 8. Bibliography Alben, L. (1996). Quality of experience: Defining the criteria for effective interaction design. Interactions, 3(3), 11-15. https://doi.org/10.1145/235008.235010. Bridger, R. S. (1995). Introduction to ergonomics. New York: McGraw-Hill. https://chooser. crossref.org/?doi=10.1201%2Fb12640. Corn, A. L. (1983). Visual function: A theoretical model for individuals with low vision. Journal of Visual Impairment & Blindness, 77(8), 373-377. https://doi.org/10.1177/0145482X8307 700801 Cushman, W. H., & Rosenberg, D. J. (1991). Human factors in product design: 14 (Advances in Human Factors/Ergonomics). Elsevier. ISBN 0444890319, 9780444890313. https://doi. org/10.1016/0003-6870(91)90234-9. Damon, A., Stoudt, H. W., & McFarland, R. A. (1966). The human body in equipment design. Harvard University Press. https://doi.org/10.1002/ajpa.1330250316. Dini, S., Ferlino, L., Gettani, A., Martinoli, C., & Ott, M. (2007). Educational software and low vision students: Evaluating accessibility factors. Universal Access in the Information Society, 6(1), 15-29. https://doi.org/10.1007/s10209-006-0056-6. Drummond, S. R., Drummond, R. S., & Dutton, G. N. (2004). Visual acuity and the ability of the visually impaired to read medication instructions. The British Journal of Ophthalmology, 88(12), 1541-2. https://doi.org/10.1136/bjo.2003.029918. Goodman, J., Langdon, P., & Clarkson, P. J. (2007). Formats for user data in inclusive design. In Stephanidis, C. (Eds) Universal access in human computer interaction. Coping with diversity: Vol 4554. (pp 117-126). Springer. https://doi.org/10.1007/978-3-540-732792_14. Jacko, J. A., Dixon, M. A., Rosa Jr, R. H., Scott, I. U., & Pappas, C. J. (1999). Visual profiles: A critical component of universal access. In: Proceedings of the SIGCHI Conf. on Human Factors in Computing Systems (pp 330-337). https://doi.org/10.1145/302979.303105. © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  17  Jacko, J. A., & Sears, A. (1998). Designing interfaces for an overlooked user group: Considering the visual profiles of partially sighted users. In: Proceedings of the 3rd int. ACM Conf. on Assistive Technologies (pp 75-77). https://doi.org/10.1145/274497.274512. Jin, B. S., & Ji. Y. G. (2010). Usability risk level evaluation for physical user interface of mobile phone. Computers in Industry, 61(4), 350-363. https://doi.org/10.1016/j.compind.2009. 12.006. Kline, D. W., & Schieber, F. (1985). Vision and aging. In J. E. Birren & K. W. Schaie (Eds.), Handbook of the psychology of aging (2nd ed., pp 296–331). Van Nostrand Reinhold Co. https://psycnet.apa.org/record/1985-98143-012. Kolko, J. (2010). Thoughts on interaction design. Massachusetts: Morgan Kaufmann. https://archive.org/details/thoughtsonintera0000kolk. Kraut, J. A., & McCabe, C. P. (1994). The problem of low vision: Definition and common problems. In D. M. Albert, F. A. Jakobiec, N. L. Robinson (Eds.) Principles and practice of ophthalmology 5 (pp 3664-6). Philadelphia: WB Saunders. Lee, J. H. (2021). Accessible home appliance design methods for disabled and elderly users (Doctoral dissertation, Dept. of Industrial Engineering. Seoul National University). https://s-space.snu.ac.kr/bitstream/10371/177380/1/000000167524.pdf Lee, J. H., Kim, Y. M., Rhiu, I., & Yun, M. H. (2021). A persona-based approach for identifying accessibility issues in elderly and disabled users’ interaction with home appliances. Applied Sciences, 11(1), 368. https://doi.org/10.3390/app11010368. Lee, J. H., You, N., Lee, J., Lee, K. J., Yun, M. H., Han, J., Joeng, Y.-J, & Lee, H. (2019). Investigation of accessibility issues for visually impaired people when using washing machines. In Bagnara, S., Tartaglia, R., Albolino, S., Alexander, T., Fujita, Y. (Eds), Proceedings of the 20th congress of the international ergonomics association (IEA 2018). Advances in intelligent systems and computing: Vol 824 (pp 1456-64). Springer Cham. https://doi.org/ 10.1007/978-3-319-96071-5_148. Lee, K. I., Jin, B. S., & Ji, Y. G. (2011). The scenario-based usability checklist development for home appliance design: A case study. Human Factors and Ergonomics in Manufacturing & Service Industries, 21(1), 67-81. https://doi.org/10.1002/hfm.20212. Mendez, J., & Mendoza, A. (2013). A conceptual framework to evaluate usability in mobile aged care applications: A health care initiative. In CONF-IRM 2013 Proceedings. 2. https://aisel.aisnet.org/confirm2013/2. Morshedzadeh, E., Ono, K., & Watanabe, M. (2016). A new model for improving user-product interaction evaluation, based on affordance and factor analysis. Bulletin of Japanese Society for the Science of Design, 62(5), 41-48. https://doi.org/10.11247/jssdj.62.5_41. Özden, A., Seheri, Ö., & Ersan, Ö. (2019). Beyaz eşya sektörü (White goods sector). Ekonomik Araştırmalar Departmanı, A&T Bank. Retrieved 15.02.2022, from https://www.atbank. com.tr/Documents/Features/ATB_beyaz-esya-sektoru_ek%C4%B1m-2019-pdf_217.PDF Pheasant, S., & Haslegrave, C. M. (2005). Bodyspace: Anthropometry, ergonomics and the design of work. CRC Press. https://doi.org/10.1201/9781315375212. Preece, J., Sharp, H., & Rogers, Y. (2015). Interaction design: Beyond human-computer interaction (4. ed.). UK: John Wiley & Sons. ISBN: 1119020751, 9781119020752. https://www.wiley. © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  18  com/en-gb/Interaction+Design%3A+Beyond+Human-Computer+Interaction%2C+6th+ Edition-p-9781119901112. Pullin, G. (2009). Design meets disability. Cambridge: MIT Press. https://mitpress.mit.edu/ 9780262516747/design-meets-disability. Sardegna, J., & Shelly, S. (2002). The encyclopedia of blindness and vision impairment. NY, USA: Infobase Publishing. Segelström, F. (2009, November). Communicating through visualizations: Service designers on visualizing user research. In DeThinking Design, ReThinking Services – First Nordic Conf. on Service Design and Service Innovation (pp 175-185). Oslo: Linköping University Electronic Press. Retrieved 10.10.2024, from https://servdes.org/pdf/2009/ segelstrom.pdf. US Department of Justice. (2010) ADA standards for accessible design. American Disability Association. Washington, DC, USA. https://www.ada.gov/law-and-regs/designstandards/2010-stds/. Vanderheiden , G. C., & Vanderheiden , K. R. (1992, January 7). Guidelines for the design of consumer products to increase their accessibility to people with disabilities or who are aging. Accessible Design of Consumer Products: University of Wisconsin System. Retrieved 10.10.2024, from https://park.org/Guests/Trace/pavilion/consumer.htm# disability. WHO. (2004). International statistical classification of diseases and health related problems ICD10: vol. 1. Geneva, Switzerland: World Health Organization. https://apps.who.int/iris /bitstream/10665/246208/1/9789241549165-V1-eng.pdf WHO. (2011) World report on disability 2011. Geneva, Switzerland. World Health Organization. https://iris.who.int/bitstream/handle/10665/44575/?sequence=4. Wood, J. M., & Troutbeck, R. J. (1994). Effect of age and visual impairment on driving and vision performance. Transp. Res. Rec., 1438, 84-90. Retrieved 20.02.2022, from https://onlinepubs.trb.org/Onlinepubs/trr/1994/1438/1438-011.pdf. © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  19  9. Appendices 9.1. Participant demographics Table 7. Low vision participant profiles Participant Age Gender Education L1 30 Female BSc L2 65 Female BSc L3 39 Male MSc L4 34 Female BSc L5 42 Female MSc L6 30 Male BSc L7 55 Male High Edu. Table 8. Normal vision participant profiles Participant Age Gender Education N1 33 Female MSc N2 35 Female BSc N3* 37 Male MSc N4* 40 Female PhD N5 33 Female BSc N6* 26 Female BSc N7* 40 Male MSc * Industrial designer © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  20  9.2. Tests of Between-Subjects effects Table 9. Tests of Between-Subjects effects for dial indicator cases Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 38260,205a 15 2550,680 11,149 ,000 ,635 Intercept 187862,223 1 187862,223 821,159 ,000 ,895 Visual Status 31255,723 1 31255,723 136,621 ,000 ,587 Case 4890,134 7 698,591 3,054 ,006 ,182 Visual Status * Case 2114,348 7 302,050 1,320 ,249 ,088 Error 21962,571 96 228,777 Total 248085,000 112 Corrected Total 60222,777 111 Dependent Variable: Time; a. R Squared = ,635 (Adjusted R Squared = ,578) Table 10. Tests of Between-Subjects effects for dial type cases Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 12447,238a 11 1131,567 10,207 ,000 ,609 Intercept 87300,762 1 87300,762 787,479 ,000 ,916 Visual Status 9945,190 1 9945,190 89,709 ,000 ,555 Case 1923,381 5 384,676 3,470 ,007 ,194 Visual Status * Case 578,667 5 115,733 1,044 ,399 ,068 Error 7982,000 72 110,861 Total 107730,000 84 Corrected Total 20429,238 83 Dependent Variable: Time; a. R Squared = ,609 (Adjusted R Squared = ,550) © Journal of Accessibility and Design for All (JACCES), Volume 15, Issue 1, 2025, ISSN: 2013-7087 DOI: https://doi.org/10.17411/jacces.v15i1.538  21  How to cite this article Aydin M. F., Yavuzcan H. G., Low vision users in graphical user interface interaction: Examining the effects of visual perception parameters on quality of experience through a display style proposal. Journal of Accessibility and Design for All, 15(1), 1–20. https://doi.org/10.17411/jacces.v15i1.538 © Journal of Accessibility and Design for All (JACCES), ISSN 2013-7087, is published by the Universitat Politècnica de Catalunya, Barcelona Tech, with the sponsorship of the ONCE Foundation for Cooperation and Social Inclusion of People with Disabilities. This issue is free of charge and available in electronic format. This work is licensed under an Attribution-NonCommercial 4.0 International Creative Commons License. Readers are allowed to read, download, copy, redistribute, print, search, or link to the full texts of the articles or use them for any other lawful purpose, giving appropriate credit. It must not be used for commercial purposes. To see the complete license contents, please visit http://creativecommons.org/licenses/by-nc/4.0/. JACCES is committed to providing accessible publications to all, regardless of technology or ability. The present document grants vital accessibility since it applies to WCAG 2.2 and PDF/UA recommendations. The evaluation tool used has been Adobe Acrobat® Accessibility Checker. If you encounter problems accessing the content of this document, you can contact us at [email protected].